From: Jesper Dangaard Brouer <brouer@redhat.com>
To: David Ahern <dsahern@gmail.com>
Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>,
Lorenzo Bianconi <lorenzo@kernel.org>,
netdev@vger.kernel.org, bpf@vger.kernel.org, davem@davemloft.net,
lorenzo.bianconi@redhat.com, echaudro@redhat.com,
sameehj@amazon.com, kuba@kernel.org, john.fastabend@gmail.com,
daniel@iogearbox.net, ast@kernel.org, shayagr@amazon.com,
David Ahern <dsahern@kernel.org>,
Ilias Apalodimas <ilias.apalodimas@linaro.org>,
brouer@redhat.com
Subject: Re: [PATCH v2 net-next 1/9] xdp: introduce mb in xdp_buff/xdp_frame
Date: Mon, 7 Sep 2020 20:02:45 +0200 [thread overview]
Message-ID: <20200907200245.0cdb63f1@carbon> (raw)
In-Reply-To: <107260d3-1fea-b582-84d3-2d092f3112b1@gmail.com>
On Fri, 4 Sep 2020 10:30:48 -0600
David Ahern <dsahern@gmail.com> wrote:
> On 9/4/20 9:59 AM, Jesper Dangaard Brouer wrote:
> >> dev_rx for example seems like it could just be the netdev
> >> index rather than a pointer or perhaps can be removed completely. I
> >> believe it is only used for 1 use case (redirects to CPUMAP); maybe that
> >> code can be refactored to handle the dev outside of xdp_frame.
> >
> > The dev_rx is needed when creating an SKB from a xdp_frame (basically
> > skb->dev = rx_dev). Yes, that is done in cpumap, but I want to
> > generalize this. The veth also creates SKBs from xdp_frame, but use
> > itself as skb->dev.
> >
> > And yes, we could save some space storing the index instead, and trade
> > space for cycles in a lookup.
>
> I think this can be managed without adding a reference to the xdp_frame.
> I'll start a separate thread on that.
>
> >>
> >> As for frame_sz, why does it need to be larger than a u16?
> >
> > Because PAGE_SIZE can be 64KiB on some archs.
> >
I also believe syzbot managed to create packets for generic-XDP with
frame_sz 128KiB, which was a bit weird (it's on my todo list to
investigate and fix).
> ok, is there any alignment requirement? can frame_sz be number of 32-bit
> words? I believe bit shifts are cheap.
No that is not possible, because some drivers and generic-XDP have a
fully dynamic frame_sz.
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
next prev parent reply other threads:[~2020-09-07 18:03 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-03 20:58 [PATCH v2 net-next 0/9] mvneta: introduce XDP multi-buffer support Lorenzo Bianconi
2020-09-03 20:58 ` [PATCH v2 net-next 1/9] xdp: introduce mb in xdp_buff/xdp_frame Lorenzo Bianconi
2020-09-04 1:07 ` Alexei Starovoitov
2020-09-04 7:19 ` Jesper Dangaard Brouer
2020-09-04 15:15 ` David Ahern
2020-09-04 15:59 ` Jesper Dangaard Brouer
2020-09-04 16:30 ` David Ahern
2020-09-07 18:02 ` Jesper Dangaard Brouer [this message]
2020-09-08 1:22 ` David Ahern
2020-09-03 20:58 ` [PATCH v2 net-next 2/9] xdp: initialize xdp_buff mb bit to 0 in all XDP drivers Lorenzo Bianconi
2020-09-04 7:35 ` Jesper Dangaard Brouer
2020-09-04 7:35 ` [Intel-wired-lan] " Jesper Dangaard Brouer
2020-09-04 7:54 ` Lorenzo Bianconi
2020-09-04 7:54 ` [Intel-wired-lan] " Lorenzo Bianconi
2020-09-03 20:58 ` [PATCH v2 net-next 3/9] net: mvneta: update mb bit before passing the xdp buffer to eBPF layer Lorenzo Bianconi
2020-09-06 7:33 ` Shay Agroskin
2020-09-06 9:05 ` Lorenzo Bianconi
2020-09-03 20:58 ` [PATCH v2 net-next 4/9] xdp: add multi-buff support to xdp_return_{buff/frame} Lorenzo Bianconi
2020-09-03 20:58 ` [PATCH v2 net-next 5/9] net: mvneta: add multi buffer support to XDP_TX Lorenzo Bianconi
2020-09-06 7:20 ` Shay Agroskin
2020-09-06 8:43 ` Lorenzo Bianconi
2020-09-03 20:58 ` [PATCH v2 net-next 6/9] bpf: helpers: add bpf_xdp_adjust_mb_header helper Lorenzo Bianconi
2020-09-04 1:09 ` Alexei Starovoitov
2020-09-04 8:07 ` Lorenzo Bianconi
2020-09-04 1:13 ` Alexei Starovoitov
2020-09-04 7:50 ` Lorenzo Bianconi
2020-09-04 13:52 ` Jesper Dangaard Brouer
2020-09-04 14:27 ` Lorenzo Bianconi
2020-09-04 6:47 ` John Fastabend
2020-09-04 9:45 ` Lorenzo Bianconi
2020-09-04 15:23 ` John Fastabend
2020-09-06 13:36 ` Lorenzo Bianconi
2020-09-08 19:57 ` John Fastabend
2020-09-08 21:31 ` Lorenzo Bianconi
2020-09-09 20:51 ` Lorenzo Bianconi
2020-09-03 20:58 ` [PATCH v2 net-next 7/9] bpf: helpers: add multibuffer support Lorenzo Bianconi
2020-09-03 21:24 ` Maciej Fijalkowski
2020-09-04 9:47 ` Lorenzo Bianconi
2020-09-03 20:58 ` [PATCH v2 net-next 8/9] samples/bpf: add bpf program that uses xdp mb helpers Lorenzo Bianconi
2020-09-03 20:58 ` [PATCH v2 net-next 9/9] net: mvneta: enable jumbo frames for XDP Lorenzo Bianconi
2020-09-04 1:08 ` [PATCH v2 net-next 0/9] mvneta: introduce XDP multi-buffer support Alexei Starovoitov
2020-09-04 7:40 ` Lorenzo Bianconi
2020-09-04 5:41 ` John Fastabend
2020-09-04 7:39 ` Lorenzo Bianconi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200907200245.0cdb63f1@carbon \
--to=brouer@redhat.com \
--cc=alexei.starovoitov@gmail.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=dsahern@gmail.com \
--cc=dsahern@kernel.org \
--cc=echaudro@redhat.com \
--cc=ilias.apalodimas@linaro.org \
--cc=john.fastabend@gmail.com \
--cc=kuba@kernel.org \
--cc=lorenzo.bianconi@redhat.com \
--cc=lorenzo@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=sameehj@amazon.com \
--cc=shayagr@amazon.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.