Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758224Ab1EBVk6 (ORCPT ); Mon, 2 May 2011 17:40:58 -0400 Received: from smtp-out.google.com ([216.239.44.51]:56664 "EHLO smtp-out.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755678Ab1EBVkz (ORCPT ); Mon, 2 May 2011 17:40:55 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=google.com; s=beta; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; b=OatCop9WISBNI56lxQnNTfW5Pu0ZN/JykmJSF4E9yRAkbmVbtV6lNg+aJ4wSJbtJVO 22YKDoclMI/nwIEc9fZQ== MIME-Version: 1.0 In-Reply-To: <1304365095.6400.20.camel@macbook.infradead.org> References: <1304365095.6400.20.camel@macbook.infradead.org> From: Mike Waychison Date: Mon, 2 May 2011 14:40:33 -0700 Message-ID: Subject: Re: linux-firmware tree To: David Woodhouse Cc: "linux-kernel@vger.kernel.org" Content-Type: text/plain; charset=ISO-8859-1 X-System-Of-Record: true Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2220 Lines: 49 On Mon, May 2, 2011 at 12:38 PM, David Woodhouse wrote: > On Mon, 2011-05-02 at 19:57 +0100, Mike Waychison wrote: >> Hi David, >> >> I'm trying to understand the purpose of the linux-firmware tree. >> Forgive me if this is a stupid question, but the only information >> about it I could find was your original post back in 2008: >> https://lwn.net/Articles/294308/ >> >> Will the Linux kernel sources always continue to have the firmware >> blobs and the linux-firmware tree is only for those users of distros >> that strip the firmware out? > > No, the legacy firmware images from the kernel tree are going to be > removed. They were only kept for backward compatibility. > > The linux-firmware tree has a lot of things that were never included in > the kernel source; no *new* firmware images are being (or at least > should be) added to the kernel source itself. > > Everyone should be using the linux-firmware tree by now. Fair enough. We haven't actually hit a case where the firmware required by any of the drivers we use isn't already present in the tree. I'm trying to figure out how we could ship firmware blobs on our systems going forward. Currently, we let firmware get installed to our install root via the modules_install target, and then package up the install root. If drivers are going to request firmware blobs from userland without them being picked up in-tree by fw-shipped-$(CONFIG_FOO), we'll need to know at build time which firmware blobs to package up. I don't see anything in the build (looking at scripts/Makefile.fwinst in particular) that allows the build to enumerate out-of-tree firmware files required by drivers. Do you have any thoughts on how this could be enumerated? (if it isn't already?) The alternative would be to ship the whole of the linux-firmware tree on every machine we have, but that seems like a bad idea as it is only going to grow with time and we don't have the MiB budget for such a thing. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/