[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20250528172507.599cce65@kernel.org>
Date: Wed, 28 May 2025 17:25:07 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Alexander Lobakin <aleksander.lobakin@...el.com>
Cc: Tony Nguyen <anthony.l.nguyen@...el.com>, <davem@...emloft.net>,
<pabeni@...hat.com>, <edumazet@...gle.com>, <andrew+netdev@...n.ch>,
<netdev@...r.kernel.org>, <maciej.fijalkowski@...el.com>,
<magnus.karlsson@...el.com>, <michal.kubiak@...el.com>,
<przemyslaw.kitszel@...el.com>, <ast@...nel.org>, <daniel@...earbox.net>,
<hawk@...nel.org>, <john.fastabend@...il.com>, <horms@...nel.org>,
<bpf@...r.kernel.org>, Mina Almasry <almasrymina@...gle.com>
Subject: Re: [PATCH net-next 01/16] libeth: convert to netmem
On Wed, 28 May 2025 16:54:39 +0200 Alexander Lobakin wrote:
> > So what happens to the packet that landed in a netmem buffer in case
> > when HDS failed? I don't see the handling.
>
> This should not happen in general, as in order to use TCP devmem, you
> need to isolate the traffic, and idpf parses all TCP frames correctly.
> If this condition is true, then napi_build_skb() will be called with the
> devmem buffer passed as head. Should I drop such packets, so that this
> would never happen?
Yes, misrouted packets can't crash the kernel.
Powered by blists - more mailing lists