From: Shay Agroskin <shayagr@amazon.com>
To: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Cc: Saeed Mahameed <saeed@kernel.org>,
Lorenzo Bianconi <lorenzo@kernel.org>,
BPF-dev-list <bpf@vger.kernel.org>,
"Network Development" <netdev@vger.kernel.org>,
"David S. Miller" <davem@davemloft.net>,
Jakub Kicinski <kuba@kernel.org>,
"Alexei Starovoitov" <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
"Jubran, Samih" <sameehj@amazon.com>,
John Fastabend <john.fastabend@gmail.com>,
David Ahern <dsahern@kernel.org>,
"Jesper Brouer" <brouer@redhat.com>,
Eelco Chaudron <echaudro@redhat.com>,
"Jason Wang" <jasowang@redhat.com>
Subject: Re: [PATCH v5 bpf-next 03/14] xdp: add xdp_shared_info data structure
Date: Mon, 21 Dec 2020 22:55:42 +0200 [thread overview]
Message-ID: <pj41zlmty6lvzl.fsf@u68c7b5b1d2d758.ant.amazon.com> (raw)
In-Reply-To: <CAJ0CqmWUJzrpOpQ01sr+e5hr1K1U4tsqEiF=FdLL--wLYpu3DA@mail.gmail.com>
Lorenzo Bianconi <lorenzo.bianconi@redhat.com> writes:
>>
>>
>> Lorenzo Bianconi <lorenzo.bianconi@redhat.com> writes:
>>
>> >> On Mon, 2020-12-07 at 17:32 +0100, Lorenzo Bianconi wrote:
>> >> > Introduce xdp_shared_info data structure to contain info
>> >> > about
>> >> > "non-linear" xdp frame. xdp_shared_info will alias
>> >> > skb_shared_info
>> >> > allowing to keep most of the frags in the same cache-line.
>> [...]
>> >>
>> >> > + u16 nr_frags;
>> >> > + u16 data_length; /* paged area length */
>> >> > + skb_frag_t frags[MAX_SKB_FRAGS];
>> >>
>> >> why MAX_SKB_FRAGS ? just use a flexible array member
>> >> skb_frag_t frags[];
>> >>
>> >> and enforce size via the n_frags and on the construction of
>> >> the
>> >> tailroom preserved buffer, which is already being done.
>> >>
>> >> this is waste of unnecessary space, at lease by definition
>> >> of
>> >> the
>> >> struct, in your use case you do:
>> >> memcpy(frag_list, xdp_sinfo->frags, sizeof(skb_frag_t) *
>> >> num_frags);
>> >> And the tailroom space was already preserved for a full
>> >> skb_shinfo.
>> >> so i don't see why you need this array to be of a fixed
>> >> MAX_SKB_FRAGS
>> >> size.
>> >
>> > In order to avoid cache-misses, xdp_shared info is built as a
>> > variable
>> > on mvneta_rx_swbm() stack and it is written to "shared_info"
>> > area only on the
>> > last fragment in mvneta_swbm_add_rx_fragment(). I used
>> > MAX_SKB_FRAGS to be
>> > aligned with skb_shared_info struct but probably we can use
>> > even
>> > a smaller value.
>> > Another approach would be to define two different struct,
>> > e.g.
>> >
>> > stuct xdp_frag_metadata {
>> > u16 nr_frags;
>> > u16 data_length; /* paged area length */
>> > };
>> >
>> > struct xdp_frags {
>> > skb_frag_t frags[MAX_SKB_FRAGS];
>> > };
>> >
>> > and then define xdp_shared_info as
>> >
>> > struct xdp_shared_info {
>> > stuct xdp_frag_metadata meta;
>> > skb_frag_t frags[];
>> > };
>> >
>> > In this way we can probably optimize the space. What do you
>> > think?
>>
>> We're still reserving ~sizeof(skb_shared_info) bytes at the end
>> of
>> the first buffer and it seems like in mvneta code you keep
>> updating all three fields (frags, nr_frags and data_length).
>> Can you explain how the space is optimized by splitting the
>> structs please?
>
> using xdp_shared_info struct we will have the first 3 fragments
> in the
> same cacheline of nr_frags while using skb_shared_info struct
> only the
> first fragment will be in the same cacheline of
> nr_frags. Moreover
> skb_shared_info has multiple fields unused by xdp.
>
> Regards,
> Lorenzo
>
Thanks for your reply. I was actually referring to your suggestion
to Saeed. Namely, defining
struct xdp_shared_info {
struct xdp_frag_metadata meta;
skb_frag_t frags[];
}
I don't see what benefits there are to this scheme compared to the
original patch
Thanks,
Shay
>>
>> >>
>> >> > +};
>> >> > +
[...]
>>
>> Saeed, the stack receives skb_shared_info when the frames are
>> passed to the stack (skb_add_rx_frag is used to add the whole
>> information to skb's shared info), and for XDP_REDIRECT use
>> case,
>> it doesn't seem like all drivers check page's tailroom for more
>> information anyway (ena doesn't at least).
>> Can you please explain what do you mean by "break the stack"?
>>
>> Thanks, Shay
>>
>> >>
>> [...]
>> >
>> >>
>>
next prev parent reply other threads:[~2020-12-21 20:58 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-07 16:32 [PATCH v5 bpf-next 00/14] mvneta: introduce XDP multi-buffer support Lorenzo Bianconi
2020-12-07 16:32 ` [PATCH v5 bpf-next 01/14] xdp: introduce mb in xdp_buff/xdp_frame Lorenzo Bianconi
2020-12-07 21:16 ` Alexander Duyck
2020-12-07 23:03 ` Saeed Mahameed
2020-12-08 3:16 ` Alexander Duyck
2020-12-08 6:49 ` Saeed Mahameed
2020-12-08 9:47 ` Jesper Dangaard Brouer
2020-12-07 16:32 ` [PATCH v5 bpf-next 02/14] xdp: initialize xdp_buff mb bit to 0 in all XDP drivers Lorenzo Bianconi
2020-12-07 21:15 ` Alexander Duyck
2020-12-07 21:37 ` Maciej Fijalkowski
2020-12-07 23:20 ` Saeed Mahameed
2020-12-08 10:31 ` Lorenzo Bianconi
2020-12-08 13:29 ` Jesper Dangaard Brouer
2020-12-07 16:32 ` [PATCH v5 bpf-next 03/14] xdp: add xdp_shared_info data structure Lorenzo Bianconi
2020-12-08 0:22 ` Saeed Mahameed
2020-12-08 11:01 ` Lorenzo Bianconi
2020-12-19 14:53 ` Shay Agroskin
2020-12-19 15:30 ` Jamal Hadi Salim
2020-12-21 9:01 ` Jesper Dangaard Brouer
2020-12-21 13:00 ` Jamal Hadi Salim
2020-12-20 17:52 ` Lorenzo Bianconi
2020-12-21 20:55 ` Shay Agroskin [this message]
2020-12-07 16:32 ` [PATCH v5 bpf-next 04/14] net: mvneta: update mb bit before passing the xdp buffer to eBPF layer Lorenzo Bianconi
2020-12-07 16:32 ` [PATCH v5 bpf-next 05/14] xdp: add multi-buff support to xdp_return_{buff/frame} Lorenzo Bianconi
2020-12-07 16:32 ` [PATCH v5 bpf-next 06/14] net: mvneta: add multi buffer support to XDP_TX Lorenzo Bianconi
2020-12-19 15:56 ` Shay Agroskin
2020-12-20 18:06 ` Lorenzo Bianconi
2020-12-07 16:32 ` [PATCH v5 bpf-next 07/14] bpf: move user_size out of bpf_test_init Lorenzo Bianconi
2020-12-07 16:32 ` [PATCH v5 bpf-next 08/14] bpf: introduce multibuff support to bpf_prog_test_run_xdp() Lorenzo Bianconi
2020-12-07 16:32 ` [PATCH v5 bpf-next 09/14] bpf: test_run: add xdp_shared_info pointer in bpf_test_finish signature Lorenzo Bianconi
2020-12-07 16:32 ` [PATCH v5 bpf-next 10/14] net: mvneta: enable jumbo frames for XDP Lorenzo Bianconi
2020-12-07 16:32 ` [PATCH v5 bpf-next 11/14] bpf: cpumap: introduce xdp multi-buff support Lorenzo Bianconi
2020-12-19 17:46 ` Shay Agroskin
2020-12-20 17:56 ` Lorenzo Bianconi
2020-12-07 16:32 ` [PATCH v5 bpf-next 12/14] bpf: add multi-buff support to the bpf_xdp_adjust_tail() API Lorenzo Bianconi
2020-12-07 16:32 ` [PATCH v5 bpf-next 13/14] bpf: add new frame_length field to the XDP ctx Lorenzo Bianconi
2020-12-08 22:17 ` Maciej Fijalkowski
2020-12-09 10:35 ` Eelco Chaudron
2020-12-09 11:10 ` Maciej Fijalkowski
2020-12-09 12:07 ` Eelco Chaudron
2020-12-15 13:28 ` Eelco Chaudron
2020-12-15 18:06 ` Maciej Fijalkowski
2020-12-16 14:08 ` Eelco Chaudron
2021-01-15 16:36 ` Eelco Chaudron
2021-01-18 16:48 ` Maciej Fijalkowski
2021-01-20 13:20 ` Eelco Chaudron
2021-02-01 16:00 ` Eelco Chaudron
2020-12-07 16:32 ` [PATCH v5 bpf-next 14/14] bpf: update xdp_adjust_tail selftest to include multi-buffer 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=pj41zlmty6lvzl.fsf@u68c7b5b1d2d758.ant.amazon.com \
--to=shayagr@amazon.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=brouer@redhat.com \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=dsahern@kernel.org \
--cc=echaudro@redhat.com \
--cc=jasowang@redhat.com \
--cc=john.fastabend@gmail.com \
--cc=kuba@kernel.org \
--cc=lorenzo.bianconi@redhat.com \
--cc=lorenzo@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=saeed@kernel.org \
--cc=sameehj@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).