From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754810Ab1CYGpr (ORCPT ); Fri, 25 Mar 2011 02:45:47 -0400 Received: from lo.gmane.org ([80.91.229.12]:42551 "EHLO lo.gmane.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754608Ab1CYGpn (ORCPT ); Fri, 25 Mar 2011 02:45:43 -0400 X-Injected-Via-Gmane: http://gmane.org/ To: linux-kernel@vger.kernel.org From: WANG Cong Subject: Re: [KDUMP] Ignore spurious IPI Date: Fri, 25 Mar 2011 06:45:25 +0000 (UTC) Message-ID: References: Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-Complaints-To: usenet@dough.gmane.org X-Gmane-NNTP-Posting-Host: 222.90.175.22 User-Agent: Pan/0.133 (House of Butterflies) Cc: kexec@lists.infradead.org Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 24 Mar 2011 08:20:32 -0600, Milton Miller wrote: > If so, just initialize the data structure earlier -- change > init_call_single_data from early_initcall to an explict call after the > per-cpu areas are initialized. > Yeah, this is a good idea, IPI related data structures should be ready before interrupts are enabled. We should not assume that there is no IPI pending before init_call_single_data gets initialized, altough this is true in non-kdump kernel. Thanks.