From: Jakub Sitnicki <jakub@cloudflare.com>
To: Cong Wang <xiyou.wangcong@gmail.com>
Cc: netdev@vger.kernel.org, bpf@vger.kernel.org,
duanxiongchun@bytedance.com, wangdongdong.6@bytedance.com,
jiang.wang@bytedance.com, Cong Wang <cong.wang@bytedance.com>,
John Fastabend <john.fastabend@gmail.com>,
Daniel Borkmann <daniel@iogearbox.net>,
Lorenz Bauer <lmb@cloudflare.com>
Subject: Re: [Patch bpf-next 03/19] skmsg: use skb ext instead of TCP_SKB_CB
Date: Fri, 05 Feb 2021 23:09:56 +0100 [thread overview]
Message-ID: <87eehu4157.fsf@cloudflare.com> (raw)
In-Reply-To: <20210203041636.38555-4-xiyou.wangcong@gmail.com>
On Wed, Feb 03, 2021 at 05:16 AM CET, Cong Wang wrote:
> From: Cong Wang <cong.wang@bytedance.com>
>
> Currently TCP_SKB_CB() is hard-coded in skmsg code, it certainly
> won't work for any other non-TCP protocols. We can move them to
> skb ext instead of playing with skb cb, which is harder to make
> correct.
>
> Of course, except ->data_end, which is used by
> sk_skb_convert_ctx_access() to adjust compile-time constant offset.
> Fortunately, we can reuse the anonymous union where the field
> 'tcp_tsorted_anchor' is and save/restore the overwritten part
> before/after a brief use.
>
> Cc: John Fastabend <john.fastabend@gmail.com>
> Cc: Daniel Borkmann <daniel@iogearbox.net>
> Cc: Jakub Sitnicki <jakub@cloudflare.com>
> Cc: Lorenz Bauer <lmb@cloudflare.com>
> Signed-off-by: Cong Wang <cong.wang@bytedance.com>
> ---
> include/linux/skbuff.h | 4 ++++
> include/linux/skmsg.h | 45 ++++++++++++++++++++++++++++++++++++++++++
> include/net/tcp.h | 25 -----------------------
> net/Kconfig | 1 +
> net/core/filter.c | 3 +--
> net/core/skbuff.c | 7 +++++++
> net/core/skmsg.c | 44 ++++++++++++++++++++++++++++-------------
> net/core/sock_map.c | 12 +++++------
> 8 files changed, 94 insertions(+), 47 deletions(-)
>
> diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
> index 46f901adf1a8..12a28268233a 100644
> --- a/include/linux/skbuff.h
> +++ b/include/linux/skbuff.h
> @@ -755,6 +755,7 @@ struct sk_buff {
> void (*destructor)(struct sk_buff *skb);
> };
> struct list_head tcp_tsorted_anchor;
> + void *data_end;
> };
I think we can avoid `data_end` by computing it in BPF with the help of
a scratch register. Similar to how we compute skb_shinfo(skb) in
bpf_convert_shinfo_access(). Something like:
static struct bpf_insn *bpf_convert_data_end_access(const struct bpf_insn *si,
struct bpf_insn *insn)
{
/* dst_reg = skb->data */
*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct sk_buff, data),
si->dst_reg, si->src_reg,
offsetof(struct sk_buff, data));
/* AX = skb->len */
*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct sk_buff, len),
BPF_REG_AX, si->src_reg,
offsetof(struct sk_buff, len));
/* dst_reg = skb->data + skb->len */
*insn++ = BPF_ALU64_REG(BPF_ADD, si->dst_reg, BPF_REG_AX);
/* AX = skb->data_len */
*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct sk_buff, data_len),
BPF_REG_AX, si->src_reg,
offsetof(struct sk_buff, data_len));
/* dst_reg = skb->data + skb->len - skb->data_len */
*insn++ = BPF_ALU64_REG(BPF_SUB, si->dst_reg, BPF_REG_AX);
return insn;
}
[...]
next prev parent reply other threads:[~2021-02-06 2:38 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-03 4:16 [Patch bpf-next 00/19] sock_map: add non-TCP and cross-protocol support Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 01/19] bpf: rename BPF_STREAM_PARSER to BPF_SOCK_MAP Cong Wang
2021-02-05 10:32 ` Jakub Sitnicki
2021-02-09 1:40 ` Cong Wang
2021-02-08 8:21 ` John Fastabend
2021-02-08 9:50 ` Lorenz Bauer
2021-02-09 1:45 ` Cong Wang
2021-02-09 6:48 ` John Fastabend
2021-02-03 4:16 ` [Patch bpf-next 02/19] skmsg: get rid of struct sk_psock_parser Cong Wang
2021-02-05 11:25 ` Jakub Sitnicki
2021-02-08 8:39 ` John Fastabend
2021-02-09 0:19 ` Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 03/19] skmsg: use skb ext instead of TCP_SKB_CB Cong Wang
2021-02-05 22:09 ` Jakub Sitnicki [this message]
2021-02-08 18:56 ` Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 04/19] sock_map: rename skb_parser and skb_verdict Cong Wang
2021-02-08 8:27 ` John Fastabend
2021-02-03 4:16 ` [Patch bpf-next 05/19] sock_map: introduce BPF_SK_SKB_VERDICT Cong Wang
2021-02-08 8:31 ` John Fastabend
2021-02-03 4:16 ` [Patch bpf-next 06/19] sock: introduce sk_prot->update_proto() Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 07/19] udp: implement ->sendmsg_locked() Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 08/19] udp: implement ->read_sock() for sockmap Cong Wang
2021-02-08 9:48 ` Lorenz Bauer
2021-02-09 1:35 ` Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 09/19] udp: add ->read_sock() and ->sendmsg_locked() to ipv6 Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 10/19] af_unix: implement ->sendmsg_locked for dgram socket Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 11/19] af_unix: implement ->read_sock() for sockmap Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 12/19] af_unix: implement ->update_proto() Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 13/19] af_unix: set TCP_ESTABLISHED for datagram sockets too Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 14/19] skmsg: extract __tcp_bpf_recvmsg() and tcp_bpf_wait_data() Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 15/19] udp: implement udp_bpf_recvmsg() for sockmap Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 16/19] af_unix: implement unix_dgram_bpf_recvmsg() Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 17/19] sock_map: update sock type checks Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 18/19] selftests/bpf: add test cases for unix and udp sockmap Cong Wang
2021-02-05 10:53 ` Jakub Sitnicki
2021-02-08 18:43 ` Cong Wang
2021-02-03 4:16 ` [Patch bpf-next 19/19] selftests/bpf: add test case for redirection between udp and unix Cong Wang
2021-02-03 17:48 ` [Patch bpf-next 00/19] sock_map: add non-TCP and cross-protocol support Alexei Starovoitov
2021-02-03 19:22 ` Cong Wang
2021-02-03 20:29 ` John Fastabend
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=87eehu4157.fsf@cloudflare.com \
--to=jakub@cloudflare.com \
--cc=bpf@vger.kernel.org \
--cc=cong.wang@bytedance.com \
--cc=daniel@iogearbox.net \
--cc=duanxiongchun@bytedance.com \
--cc=jiang.wang@bytedance.com \
--cc=john.fastabend@gmail.com \
--cc=lmb@cloudflare.com \
--cc=netdev@vger.kernel.org \
--cc=wangdongdong.6@bytedance.com \
--cc=xiyou.wangcong@gmail.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).