From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH 0/5] kvm: qemu: virtio_net: add support for mergeable rx buffers Date: Thu, 27 Nov 2008 14:45:59 +0200 Message-ID: <492E9687.7030501@redhat.com> References: <1223991840.11098.29.camel@blaa> <1227711038-21942-1-git-send-email-markmc@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: kvm@vger.kernel.org, Rusty Russell , Herbert Xu , Anthony Liguori To: Mark McLoughlin Return-path: Received: from mx2.redhat.com ([66.187.237.31]:43544 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752636AbYK0MqH (ORCPT ); Thu, 27 Nov 2008 07:46:07 -0500 In-Reply-To: <1227711038-21942-1-git-send-email-markmc@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: Mark McLoughlin wrote: > Hi, > The mergeable receive buffers scheme has been in the Linux > virtio_net driver in net-next-2.6 for a while now, so it's probably > safe to assume what's there now is the final iteration of the ABI. > > The following patches implement support for the scheme in > KVM. Only the 5/5 patch contains functional changes. > Applied all, thanks. > Changes since last time: > > + We now peak at how much buffer space is actually available in the > ring and no longer assume that buffers are at least 4k > > + The new header is now 'struct virtio_net_hdr_mrg_rxbuf' rather > than 'struct virtio_net_hdr2' > Does the updated driver support both? > + The new header no longer is padded to 32 bytes > Pity. -- error compiling committee.c: too many arguments to function