From mboxrd@z Thu Jan 1 00:00:00 1970 From: Brice Goglin Subject: Re: [PATCH 4/6] myri10ge - First half of the driver Date: Mon, 15 May 2006 19:39:47 +0200 Message-ID: <4468BCE3.3010005@myri.com> References: <446259A0.8050504@myri.com> <20060510231347.GC25334@electric-eye.fr.zoreil.com> <4463CE88.20301@myri.com> <1147712555.27252.269.camel@mindpipe> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Francois Romieu , netdev@vger.kernel.org, LKML , "Andrew J. Gallatin" Return-path: Received: from h-66-166-126-70.lsanca54.covad.net ([66.166.126.70]:2964 "EHLO myri.com") by vger.kernel.org with ESMTP id S1751615AbWEORkH (ORCPT ); Mon, 15 May 2006 13:40:07 -0400 To: Lee Revell In-Reply-To: <1147712555.27252.269.camel@mindpipe> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Lee Revell wrote: > On Fri, 2006-05-12 at 01:53 +0200, Brice Goglin wrote: > >> Francois Romieu wrote: >> >>>> + spin_lock(&mgp->cmd_lock); >>>> + response->result = 0xffffffff; >>>> + mb(); >>>> + myri10ge_pio_copy((void __iomem *) cmd_addr, buf, sizeof (*buf)); >>>> + >>>> + /* wait up to 2 seconds */ >>>> >>>> >>> You must not hold a spinlock for up to 2 seconds. >>> >>> >> We are working on reducing the delay to about 15ms. It only occurs when >> the driver is loaded or the link brought up. >> > > I think 15ms is quite a long time to hold a spinlock also - most > spinlocks in the kernel are held for less than 500 microseconds. > > Can't you use a mutex? > It looks like rtnl_lock protects us here. We are working on it. Brice