netdev - Re: [PATCH net-next v2 2/2] net/mlx5: Avoid copying payload to the skb's linear part

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] [day] [month] [year] [list]

Message-ID: <CADg4-L__JBO8j4tVnG8-DxCdmbUWQQqXjgb1W3qb4=29F1c=HA@mail.gmail.com>
Date: Thu, 21 Aug 2025 15:59:29 -0700
From: Christoph Paasch <cpaasch@...nai.com>
To: Jakub Kicinski <kuba@...nel.org>
Cc: Dragos Tatulea <dtatulea@...dia.com>, Saeed Mahameed <saeedm@...dia.com>, 
	Leon Romanovsky <leon@...nel.org>, Tariq Toukan <tariqt@...dia.com>, Mark Bloch <mbloch@...dia.com>, 
	Andrew Lunn <andrew+netdev@...n.ch>, "David S. Miller" <davem@...emloft.net>, 
	Eric Dumazet <edumazet@...gle.com>, Paolo Abeni <pabeni@...hat.com>, 
	Alexander Lobakin <aleksander.lobakin@...el.com>, Gal Pressman <gal@...dia.com>, 
	linux-rdma@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [PATCH net-next v2 2/2] net/mlx5: Avoid copying payload to the
 skb's linear part

On Tue, Aug 19, 2025 at 5:15 PM Jakub Kicinski <kuba@...nel.org> wrote:
>
> On Tue, 19 Aug 2025 09:58:54 +0000 Dragos Tatulea wrote:
> > > @@ -2009,10 +2040,14 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
> > >     u32 linear_frame_sz;
> > >     u16 linear_data_len;
> > >     u16 linear_hr;
> > > +   u16 headlen;
> > >     void *va;
> > >
> > >     prog = rcu_dereference(rq->xdp_prog);
> > >
> > > +   headlen = min3(mlx5e_cqe_estimate_hdr_len(cqe), cqe_bcnt,
> > > +                  (u16)MLX5E_RX_MAX_HEAD);
> > > +
> > How about keeping the old calculation for XDP and do this one for
> > non-xdp in the following if/else block?
> >
> > This way XDP perf will not be impacted by the extra call to
> > mlx5e_cqe_estimate_hdr_len().
>
> Perhaps move it further down for XDP?
> Ideally attaching a program which returns XDP_PASS shouldn't impact
> normal TCP perf.

Yes, makes sense!

Will do that and resubmit.


Thanks,
Christoph