From mboxrd@z Thu Jan 1 00:00:00 1970 From: Darek Marcinkiewicz Subject: Re: [PATCH v5 1/1] Driver for Beckhoff CX5020 EtherCAT master module. Date: Sun, 4 May 2014 23:41:17 +0200 Message-ID: <20140504214116.GI1156@newterm.pl> References: <20140502175242.GB1156@newterm.pl> <20140503114029.GA19315@electric-eye.fr.zoreil.com> <20140504110413.GE1156@newterm.pl> <20140504184351.GA4471@electric-eye.fr.zoreil.com> <20140504194624.GG1156@newterm.pl> <20140504211928.GB4471@electric-eye.fr.zoreil.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: davem@davemloft.net, netdev@vger.kernel.org, linux-kernel@vger.kernel.org To: Francois Romieu Return-path: Content-Disposition: inline In-Reply-To: <20140504211928.GB4471@electric-eye.fr.zoreil.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Sun, May 04, 2014 at 11:19:28PM +0200, Francois Romieu wrote: > Darek Marcinkiewicz : > > On Sun, May 04, 2014 at 08:43:51PM +0200, Francois Romieu wrote: > [...] > > > Regarding tx_dnext updates, you may add a short notice in ec_bhf_start_xmit > > > and ec_bhf_process_tx explaining that the periodic poller will somehow end > > > working with the right value, whence no (smp_)barrier at all. > > > > > Hmm, good point. I am not really sure that it is not a race. So, I've added > > memory barriers for the case when the tx ring becomes full. > > Without memory barrier, the hrtimer poller may be wrong but 1) it will > always be pessimistic and 2) it won't last. It would be sloppy though. > > Did you have some time to test the latest version ? > Yes, I have this code continuously running and it looks good. This is a regular application that I have running now, so not all interesting edge cases might have been exercised, though. -- DM