From mboxrd@z Thu Jan 1 00:00:00 1970 From: Francois Romieu Subject: Re: [PATCH] net/r8169: Remove the firmware of RTL8111D Date: Mon, 6 Dec 2010 23:04:40 +0100 Message-ID: <20101206220440.GA3853@electric-eye.fr.zoreil.com> References: <1291180966-3124-1-git-send-email-hayeswang@realtek.com> <20101201080732.GA3234@electric-eye.fr.zoreil.com> <1291610284.8025.154.camel@localhost> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Hayes Wang , netdev@vger.kernel.org, linux-kernel@vger.kernel.org To: Ben Hutchings Return-path: Received: from violet.fr.zoreil.com ([92.243.8.30]:36717 "EHLO violet.fr.zoreil.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753384Ab0LFWEw (ORCPT ); Mon, 6 Dec 2010 17:04:52 -0500 Content-Disposition: inline In-Reply-To: <1291610284.8025.154.camel@localhost> Sender: netdev-owner@vger.kernel.org List-ID: Ben Hutchings : [...] > I have reports that the RTL8111D/8168D hardware can work without this > firmware, possibly with limited functionality or compatibility. (I > assume it is a patch for the original firmware loaded from ROM or > flash.) So I feel obliged to give users the choice to use the driver > without installing non-free firmware. If you want to make the warning > on failure of request_firmware() more explicit then that's OK with me. I'll make it more explicit regarding what it loads, yes. > If you make the failure fatal I will probably revert that in Debian. I have no technical argument against yours to make the failure fatal yet: even Realtek's wording is rather soft. I will reconsider it if things turn ugly. Btw, it is not too hard to build a firmware-of-freedom binary : all zero should be fine here. -- Ueimor