Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261277AbTHXQc1 (ORCPT ); Sun, 24 Aug 2003 12:32:27 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261269AbTHXQc1 (ORCPT ); Sun, 24 Aug 2003 12:32:27 -0400 Received: from dbl.q-ag.de ([80.146.160.66]:62095 "EHLO dbl.q-ag.de") by vger.kernel.org with ESMTP id S261277AbTHXQb6 (ORCPT ); Sun, 24 Aug 2003 12:31:58 -0400 Message-ID: <3F48E870.2070205@colorfullife.com> Date: Sun, 24 Aug 2003 18:31:44 +0200 From: Manfred Spraul User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030701 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Vinay K Nallamothu CC: linux-kernel@vger.kernel.org Subject: Re: [PATCH 2.6.0-test4][NET] ni5010.c: remove cli/sti Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 895 Lines: 25 Vinay wrote: >drivers/net/ni5010.c: >This patch replaces cli/sti with spinlocks. Compiles fine though >untested. > > I don't have the hardware either, but the patch seems to be wrong: The cli() call was in hardware_send_packet(). I assume that this function should be synchronized against concurrent interrupts - both functions access the hardware. Due to the dev_xmit_lock, the networking core already guarantees that hardware_send_packet is never reentered. cli() provided protection against concurrent interrupts on other cpus, spin_lock_irqsave() doesn't. Could you add a spin_lock()/spin_unlock() into ni5010_interrupt? -- Manfred - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/