From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sowmini Varadhan Subject: Re: [PATCH net-next] af_packet: Provide a TPACKET_V2 compatible Tx path for TPACKET_V3 Date: Sat, 31 Dec 2016 07:21:38 -0500 Message-ID: <20161231122138.GA5627@oracle.com> References: <1483116853-45769-1-git-send-email-sowmini.varadhan@oracle.com> <20161230230316.GB31800@oracle.com> <20161231004826.GC31800@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Network Development , Daniel Borkmann , Willem de Bruijn , David Miller To: Willem de Bruijn Return-path: Received: from aserp1040.oracle.com ([141.146.126.69]:18394 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752703AbcLaMWH (ORCPT ); Sat, 31 Dec 2016 07:22:07 -0500 Content-Disposition: inline In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On (12/30/16 23:59), Willem de Bruijn wrote: > > Actually I'm not averse to looking at extensions (or at least, > > place-holders) to allow variable sized slots- do you have any > > suggestions? > > It is probably enough to just enforce that tp_next_offset is always > sane. Either that it points to the start of the next packet, even > though that currently can also be inferred from frame_size. Or, that > it is always zero unless the ring is to be interpreted as holding > variable sized frames. If the field is non-zero, drop the packet. Ok, let me add this to patchv2, along with extra test cases. --Sowmini