From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?iso-8859-2?Q?Micha=B3_Miros=B3aw?= Subject: Re: [PATCH] net: tun: convert to hw_features Date: Thu, 21 Apr 2011 01:15:03 +0200 Message-ID: <20110420231503.GA20775@rere.qmqm.pl> References: <20110419161310.7508513909@rere.qmqm.pl> <871v0xip3j.fsf@rustcorp.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-2 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org To: Rusty Russell Return-path: Received: from rere.qmqm.pl ([89.167.52.164]:45581 "EHLO rere.qmqm.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752292Ab1DTXPF (ORCPT ); Wed, 20 Apr 2011 19:15:05 -0400 Content-Disposition: inline In-Reply-To: <871v0xip3j.fsf@rustcorp.com.au> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, Apr 20, 2011 at 12:52:24PM +0930, Rusty Russell wrote: > On Tue, 19 Apr 2011 18:13:10 +0200 (CEST), Micha=B3 Miros=B3aw wrote: > > This changes offload setting behaviour to what I think is correct: > > - offloads set via ethtool mean what admin wants to use (by defaul= t > > he wants 'em all) > > - offloads set via ioctl() mean what userspace is expecting to get > > (this limits which admin wishes are granted) > > - TUN_NOCHECKSUM is ignored, as it might cause broken packets when > > forwarded (ip_summed =3D=3D CHECKSUM_UNNECESSARY means that chec= ksum > > was verified, not that it can be ignored) > >=20 > > If TUN_NOCHECKSUM is implemented, it should set skb->csum_* and > > skb->ip_summed (=3D CHECKSUM_PARTIAL) for known protocols and let o= thers > > be verified by kernel when necessary. > >=20 > > TUN_NOCHECKSUM handling was introduced by commit > > f43798c27684ab925adde7d8acc34c78c6e50df8: > >=20 > > tun: Allow GSO using virtio_net_hdr > Err, not in my git tree! It predates git in fact. Ah! Sorry! Your patch just moved the buggy line around. > Since tap requires privs, I wouldn't worry about invalid packets too > much. I prefer correctness over uncertainty. ;) Best Regards, Micha=B3 Miros=B3aw