[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAF=yD-LFT+qqctdntnEOnhpwG-1+YPt=D4-5Mm0R2A9cMmoOFw@mail.gmail.com>
Date: Thu, 16 Nov 2023 16:23:57 -0500
From: Willem de Bruijn <willemdebruijn.kernel@...il.com>
To: Eric Dumazet <edumazet@...gle.com>
Cc: "David S . Miller" <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>, netdev@...r.kernel.org, eric.dumazet@...il.com,
Bailey Forrest <bcf@...gle.com>, Willem de Bruijn <willemb@...gle.com>,
Jeroen de Borst <jeroendb@...gle.com>, Praveen Kaligineedi <pkaligineedi@...gle.com>,
Shailend Chand <shailend@...gle.com>, Ziwei Xiao <ziweixiao@...gle.com>
Subject: Re: [PATCH net] gve: add gve_features_check()
On Thu, Nov 16, 2023 at 3:57 AM Eric Dumazet <edumazet@...gle.com> wrote:
>
> It is suboptimal to attempt skb linearization from ndo_start_xmit()
> if a gso skb has pathological layout, or if host stack does not have
> access to the payload (TCP direct). Linearization of large skbs
> can also fail under memory pressure.
>
> We should instead have an ndo_features_check() so that we can
> fallback to GSO, which is supported even for TCP direct,
> and generally much more efficient (no payload copy).
>
> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
> Cc: Bailey Forrest <bcf@...gle.com>
> Cc: Willem de Bruijn <willemb@...gle.com>
> Cc: Jeroen de Borst <jeroendb@...gle.com>
> Cc: Praveen Kaligineedi <pkaligineedi@...gle.com>
> Cc: Shailend Chand <shailend@...gle.com>
> Cc: Ziwei Xiao <ziweixiao@...gle.com>
Reviewed-by: Willem de Bruijn <willemb@...gle.com>
Powered by blists - more mailing lists