From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH net-next] virtio-net: invoke zerocopy callback on xmit path if no tx napi Date: Tue, 22 Aug 2017 20:55:56 +0300 Message-ID: <20170822204015-mutt-send-email-mst@kernel.org> References: <20170819063854.27010-1-den@klaipeden.com> <5352c98a-fa48-fcf9-c062-9986a317a1b0@redhat.com> <64d451ae-9944-e978-5a05-54bb1a62aaad@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Willem de Bruijn , Koichiro Den , virtualization@lists.linux-foundation.org, Network Development To: Jason Wang Return-path: Received: from mx1.redhat.com ([209.132.183.28]:54070 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751915AbdHVRz6 (ORCPT ); Tue, 22 Aug 2017 13:55:58 -0400 Content-Disposition: inline In-Reply-To: <64d451ae-9944-e978-5a05-54bb1a62aaad@redhat.com> Sender: netdev-owner@vger.kernel.org List-ID: On Tue, Aug 22, 2017 at 10:50:41AM +0800, Jason Wang wrote: > > Perhaps the descriptor pool should also be > > revised to allow out of order completions. Then there is no need to > > copy zerocopy packets whenever they may experience delay. > > Yes, but as replied in the referenced thread, windows driver may treat out > of order completion as a bug. That would be a windows driver bug then, but I don't think it makes this assumption. What the referenced thread (https://patchwork.kernel.org/patch/3787671/) is saying is that host must use any buffers made available on a tx vq within a reasonable timeframe otherwise windows guests panic. Ideally we would detect that a packet is actually experiencing delay and trigger the copy at that point e.g. by calling skb_linearize. But it isn't easy to track these packets though and even harder to do a data copy without races. Which reminds me that skb_linearize in net core seems to be fundamentally racy - I suspect that if skb is cloned, and someone is trying to use the shared frags while another thread calls skb_linearize, we get some use after free bugs which likely mostly go undetected because the corrupted packets mostly go on wire and get dropped by checksum code. -- MST