[Intel-wired-lan] [PATCH v4 net-next 2/9] i40e: respect metadata on XSK Rx to skb
Alexander Lobakin
alexandr.lobakin at intel.com
Fri Dec 10 11:08:48 UTC 2021
From: Tony Nguyen <anthony.l.nguyen at intel.com>
Date: Thu, 9 Dec 2021 18:50:07 +0000
> On Thu, 2021-12-09 at 18:38 +0100, Alexander Lobakin wrote:
> > From: Jesper Dangaard Brouer <jbrouer at redhat.com>
> > Date: Thu, 9 Dec 2021 09:27:37 +0100
> >
> > > On 08/12/2021 15.06, Alexander Lobakin wrote:
> > > > For now, if the XDP prog returns XDP_PASS on XSK, the metadata
> > > > will
> > > > be lost as it doesn't get copied to the skb.
> > >
> > > I have an urge to add a newline here, when reading this, as IMHO it
> > > is a
> > > paragraph with the problem statement.
> > >
> > > > Copy it along with the frame headers. Account its size on skb
> > > > allocation, and when copying just treat it as a part of the frame
> > > > and do a pull after to "move" it to the "reserved" zone.
> > >
> > > Also newline here, as next paragraph are some extra details, you
> > > felt a
> > > need to explain to the reader.
> > >
> > > > net_prefetch() xdp->data_meta and align the copy size to speed-up
> > > > memcpy() a little and better match i40e_costruct_skb().
> > > ^^^^^^xx^^^^^^^^^
> > >
> > > commit messages.
> >
> > Oh gosh, I thought I don't have attention deficit. Thanks, maybe
> > Tony will fix it for me or I could send a follow-up (or resend if
> > needed, I saw those were already applied to dev-queue).
>
> If there's no need for follow-ups beyond this change, I'll fix it up.
The rest is fine, thank you!
> Thanks,
> Tony
>
> > > --Jesper
> >
> > Al
Al
More information about the Intel-wired-lan
mailing list