[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220214074824.370d7ae6@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
Date: Mon, 14 Feb 2022 07:48:24 -0800
From: Jakub Kicinski <kuba@...nel.org>
To: Lorenzo Bianconi <lorenzo@...nel.org>
Cc: bpf@...r.kernel.org, netdev@...r.kernel.org, davem@...emloft.net,
ast@...nel.org, daniel@...earbox.net, brouer@...hat.com,
toke@...hat.com, pabeni@...hat.com, echaudro@...hat.com,
lorenzo.bianconi@...hat.com, toshiaki.makita1@...il.com,
andrii@...nel.org
Subject: Re: [PATCH bpf-next 2/3] veth: rework veth_xdp_rcv_skb in order to
accept non-linear skb
On Sat, 12 Feb 2022 12:03:49 +0100 Lorenzo Bianconi wrote:
> On Feb 11, Jakub Kicinski wrote:
> > On Fri, 11 Feb 2022 02:20:31 +0100 Lorenzo Bianconi wrote:
> > > + if (skb_shared(skb) || skb_head_is_locked(skb)) {
> >
> > Is this sufficient to guarantee that the frags can be written?
> > skb_cow_data() tells a different story.
>
> Do you mean to consider paged part of the skb always not writable, right?
> In other words, we should check something like:
>
> if (skb_shared(skb) || skb_head_is_locked(skb) ||
> skb_shinfo(skb)->nr_frags) {
> ...
> }
Yes, we do have skb_has_shared_frag() but IDK if it guarantees frags
are writable :S
Powered by blists - more mailing lists