From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH net-next] cxgb4: set skb->rxhash Date: Tue, 27 Apr 2010 15:47:41 -0700 (PDT) Message-ID: <20100427.154741.172557759.davem@davemloft.net> References: <1272406825-11615-1-git-send-email-dm@chelsio.com> <20100427.154338.223483726.davem@davemloft.net> <4BD76955.1030104@chelsio.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org To: dm@chelsio.com Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:46822 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757454Ab0D0Wrf (ORCPT ); Tue, 27 Apr 2010 18:47:35 -0400 In-Reply-To: <4BD76955.1030104@chelsio.com> Sender: netdev-owner@vger.kernel.org List-ID: From: Dimitris Michailidis Date: Tue, 27 Apr 2010 15:46:45 -0700 > David Miller wrote: >> From: Dimitris Michailidis >> Date: Tue, 27 Apr 2010 15:20:25 -0700 >> >>> Implement the ->set_flags ethtool method to control NETIF_F_RXHASH and >>> set skb->rxhash to the HW calculated hash accordingly. >>> >>> Signed-off-by: Dimitris Michailidis >> Does the CXGB4 hash the ports for non-TCP packets? >> If not, you will need to check the packet type (preferrably using >> RX descriptor information rather than actually parsing the packet) >> and only record the ->rxhash for TCP packets. >> > > It does, optionally. Someday I need to implement ->set_rxnfc to > expose the few available options but it's on by default. Great, if you could respin this patch and ditch the ntohl() as Eric suggested I'll apply it. Thanks!