From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [net-next PATCH 1/3] vxge: always enable hardware time stamp Date: Tue, 12 Apr 2011 11:01:51 -0700 (PDT) Message-ID: <20110412.110151.71117421.davem@davemloft.net> References: <1302297083-22665-1-git-send-email-jdmason@kudzu.us> <20110410.185845.70188298.davem@davemloft.net> <20110412153604.GA1433@kudzu.us> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org To: jdmason@kudzu.us Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:36657 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758080Ab1DLSC1 (ORCPT ); Tue, 12 Apr 2011 14:02:27 -0400 In-Reply-To: <20110412153604.GA1433@kudzu.us> Sender: netdev-owner@vger.kernel.org List-ID: From: Jon Mason Date: Tue, 12 Apr 2011 10:36:06 -0500 > On Sun, Apr 10, 2011 at 06:58:45PM -0700, David Miller wrote: >> From: Jon Mason >> Date: Fri, 8 Apr 2011 16:11:21 -0500 >> >> > Hardware time stamp calculation can only be enabled by the privileged >> > function. Enable it always by default and simply use the ethtool >> > interface to set a flag to indicate whether or not the respective >> > function driver should indicate the timestamp along with the received >> > packet. >> > >> > Also, make certain fields in vxge_hw_device_config bit-fields to reduce >> > the size of the struct. >> > >> > Signed-off-by: Jon Mason >> >> Doesn't this have some performance or latency impact? > > It is all done in hardware by replacing the CRC with the HWTS value. > So, no perf or latency issues there. It still only handles the HWTS > in receive if it is enabled in software via the ioctl. Ok, thanks for the clarification, I'll apply this patch set.