From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hauke Mehrtens Subject: Re: [PATCH 2/4] bgmac: write mac address to hardware in ndo_set_mac_address Date: Wed, 06 Feb 2013 16:17:33 +0100 Message-ID: <5112740D.2000402@hauke-m.de> References: <1360161900-2330-1-git-send-email-hauke@hauke-m.de> <1360161900-2330-3-git-send-email-hauke@hauke-m.de> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: davem@davemloft.net, netdev@vger.kernel.org To: =?UTF-8?B?UmFmYcWCIE1pxYJlY2tp?= Return-path: Received: from server19320154104.serverpool.info ([193.201.54.104]:34831 "EHLO hauke-m.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755801Ab3BFPRl (ORCPT ); Wed, 6 Feb 2013 10:17:41 -0500 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On 02/06/2013 03:57 PM, Rafa=C5=82 Mi=C5=82ecki wrote: > 2013/2/6 Hauke Mehrtens : >> The generic implementation just changes the netdev struct and does n= ot >> write the new mac address to the hardware or issues some command to = do >> so. >=20 > That's because generic "eth_mac_addr" allows changing MAC only when > interface is down. And we "upload" MAC address during bringing > interface up. So there wasn't any bug, it was just working that way := ) >=20 > One question: how you tested if changing MAC with keeping interface u= p > works correctly? I just don't know if simple writes to > BGMAC_MACADDR_HIGH and BGMAC_MACADDR_LOW are enough. >=20 > If it works for you (changing MAC without stopping interface), I'm OK= with that. >=20 Yes that works in promisc and non promisc mode, I did this while pingin= g the device and I had no package lose. Hauke