From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arend van Spriel Subject: Re: [PATCH 01/14] brcmfmac: add delay before unregistering the network device Date: Wed, 26 Feb 2014 12:34:09 +0100 Message-ID: <530DD131.50205@broadcom.com> References: <1393356639-19169-1-git-send-email-arend@broadcom.com> <1393356639-19169-2-git-send-email-arend@broadcom.com> (sfid-20140225_203048_842452_A39499B6) <1393358376.4170.22.camel@jlt4.sipsolutions.net> <530DAECD.8010405@broadcom.com> <1393406278.4133.4.camel@jlt4.sipsolutions.net> <530DB3BA.20106@broadcom.com> <1393407473.4133.5.camel@jlt4.sipsolutions.net> <530DC56E.9040202@broadcom.com> <1393413044.4133.6.camel@jlt4.sipsolutions.net> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: "John W. Linville" , linux-wireless , David Miller , To: Johannes Berg Return-path: Received: from mail-gw2-out.broadcom.com ([216.31.210.63]:61351 "EHLO mail-gw2-out.broadcom.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751202AbaBZLeO (ORCPT ); Wed, 26 Feb 2014 06:34:14 -0500 In-Reply-To: <1393413044.4133.6.camel@jlt4.sipsolutions.net> Sender: netdev-owner@vger.kernel.org List-ID: On 02/26/2014 12:10 PM, Johannes Berg wrote: > On Wed, 2014-02-26 at 11:43 +0100, Arend van Spriel wrote: > >> Actually, this was not the reason for the doing the msleep(). The >> problem was that wpa_supplicant received RTM_DELLINK, before getting the >> NL08211_DISCONNECT. This resulted in RTM_DELLINK, immediately followed >> by RTM_ADDLINK, followed by NL80211_DISCONNECT. > > I'm not sure this isn't still possible? I don't know when DELLINK is > sent. Given the behaviour I would say before the netdev notifier, which may be considered wrong. It depends whether the notifier was intended for the kind of thing that cfg80211 is doing, ie. send a netlink event to user-space? Gr. AvS