From mboxrd@z Thu Jan 1 00:00:00 1970 From: Divy Le Ray Subject: Re: [PATCH] cxgb4vf: fix crash due to manipulating queues before registration Date: Mon, 01 Nov 2010 14:02:56 -0700 Message-ID: <4CCF2B00.3060602@chelsio.com> References: <1288297007-663-1-git-send-email-leedom@chelsio.com> <20101028.132122.245402585.davem@davemloft.net> <8A71B368A89016469F72CD08050AD334088DA49E@maui.asicdesigners.com> <20101029.130559.112585228.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: dm@chelsio.com, leedom@chelsio.com, netdev@vger.kernel.org To: David Miller Return-path: Received: from stargate.chelsio.com ([67.207.112.58]:6352 "EHLO stargate.chelsio.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754323Ab0KAVBd (ORCPT ); Mon, 1 Nov 2010 17:01:33 -0400 In-Reply-To: <20101029.130559.112585228.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: On 10/29/2010 01:05 PM, David Miller wrote: > From: "Dimitrios Michailidis" > Date: Fri, 29 Oct 2010 00:36:22 -0700 > >> Further, I believe moving the call after register_netdev is buggy as >> open can be called after registration and it can clash with the >> queue stopping. It seems then that these netif_tx_stop_all_queues >> calls have to go now. > This is a good explanation of why no driver should be touching the > queue state before the first ->open() call. I'm sending a series of 3 patches fixing cxgb3, cxgb4, cxgb4vf. Cheers, Divy