From mboxrd@z Thu Jan 1 00:00:00 1970 From: Constantin Baranov Subject: Re: [PATCH 2/2] niu: improve ethtool support for gigabit copper cards Date: Tue, 17 Feb 2009 18:43:49 +0400 Message-ID: <20090217184349.3c735c58.baranov@mercdev.com> References: <20090217154334.97837477.baranov@mercdev.com> <1234878837.3183.4.camel@achroite> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: "David S. Miller" , netdev@vger.kernel.org, linux-kernel@vger.kernel.org To: Ben Hutchings Return-path: Received: from titan.dewahost.net ([205.234.111.6]:48913 "EHLO titan.dewahost.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751790AbZBQOn5 (ORCPT ); Tue, 17 Feb 2009 09:43:57 -0500 In-Reply-To: <1234878837.3183.4.camel@achroite> Sender: netdev-owner@vger.kernel.org List-ID: Ben Hutchings wrote: > On Tue, 2009-02-17 at 15:43 +0400, Constantin Baranov wrote: > > Introduced support for link speed and duplex setting (ethtool -s), > > link advertising parameters and autonegotiation (ethtool -r): > [...] > > I haven't looked closely at this, but is there anything here that can't > be handled by mii_get_settings() and other functions in the mii driver? mii_ethtool_gset() reports fixed set of supported modes and mii_ethtool_sset() doesn't check whether requested mode is supported when autonegotiation is disabled. Also mii module declares mdio_write() operation as returning void, thus mii routines will ignore errors that may occur. Although all of these is not so critical I decided not to use mii module. -- Constantin Baranov, Mercury Development, LLC