From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Greear Subject: Re: [PATCH] New version of net timestamp optimization Date: Wed, 30 Jun 2004 11:46:38 -0700 Sender: netdev-bounce@oss.sgi.com Message-ID: <40E30A8E.6000507@candelatech.com> References: <20040406182941.7e759f4c.ak@suse.de> <1081340659.15893.117.camel@jzny.localdomain> <20040407143703.6af5bb71.ak@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: hadi@cyberus.ca, netdev@oss.sgi.com Return-path: To: Andi Kleen In-Reply-To: <20040407143703.6af5bb71.ak@suse.de> Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org Andi Kleen wrote: > On 07 Apr 2004 08:24:19 -0400 > jamal wrote: > > >>The patch does look clean and attractive. I havent had time toi play >>with it myself - but given the way gettimeofday has been torturing me >>i will at some point. Also if you have time a 2.4.x patch would be > > > Thanks. All feedback I received so far was positive. I hope DaveM will > merge it soon. I don't have plans to work on 2.4 anymore. I finally got around to working with 2.6, and I ran into the new timestamp code. I would like to be able to have timestamps even though I do not have a socket structure (I am using a kernel module to grab all frames)... How would you feel about either exporting the netstamp_needed variable or offering a method that explicitly incremented or decremented the use count? Thanks, Ben > > -Andi > -- Ben Greear Candela Technologies Inc http://www.candelatech.com