netdev - Re: [PATCH net-next 2/4] sfc: implement encap TSO on EF100

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CA+FuTSdk-UZ92VdpWTAx87xnzhsDKcWfVOOwG_B16HdAuP7PQA@mail.gmail.com>
Date:   Fri, 30 Oct 2020 12:26:43 -0400
From:   Willem de Bruijn <willemdebruijn.kernel@...il.com>
To:     Edward Cree <ecree@...arflare.com>
Cc:     Willem de Bruijn <willemdebruijn.kernel@...il.com>,
        linux-net-drivers@...arflare.com, Jakub Kicinski <kuba@...nel.org>,
        David Miller <davem@...emloft.net>,
        Network Development <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next 2/4] sfc: implement encap TSO on EF100

On Fri, Oct 30, 2020 at 12:16 PM Edward Cree <ecree@...arflare.com> wrote:
>
> On 30/10/2020 15:49, Willem de Bruijn wrote:
> > On Wed, Oct 28, 2020 at 9:39 PM Edward Cree <ecree@...arflare.com> wrote:
> >> +                             ESF_GZ_TX_TSO_ED_OUTER_UDP_LEN, encap && !gso_partial,
> >
> > This is a boolean field to signal whether the NIC needs to fix up the
> > udp length field ?
> Yes.

Thanks

> > Which in the case of GSO_PARTIAL has already been resolved by the gso
> > layer (in __skb_udp_tunnel_segment).
> Indeed.
>
> > Just curious, is this ever expected to be true? Not based on current
> > advertised features, right?
> As I mentioned in the patch description and cover letter, I'm not
>  entirely certain.  I don't _think_ the stack will ever give us an
>  encap skb without GSO_PARTIAL with the features we've advertised,

That's my understanding too.

>  but since the hardware supports it I thought it better to handle
>  that case anyway, just in case I'm mistaken.

Then you could (as follow-up) advertise without GSO_PARTIAL and avoid
the whole transition through the gso layer?