From: John Fastabend <john.fastabend@gmail.com>
To: cong.wang@bytedance.com, jakub@cloudflare.com,
daniel@iogearbox.net, lmb@isovalent.com, edumazet@google.com
Cc: john.fastabend@gmail.com, bpf@vger.kernel.org,
netdev@vger.kernel.org, ast@kernel.org, andrii@kernel.org,
will@isovalent.com
Subject: [PATCH bpf v2 04/12] bpf: sockmap, handle fin correctly
Date: Mon, 27 Mar 2023 10:54:38 -0700 [thread overview]
Message-ID: <20230327175446.98151-5-john.fastabend@gmail.com> (raw)
In-Reply-To: <20230327175446.98151-1-john.fastabend@gmail.com>
The sockmap code is returning EAGAIN after a FIN packet is received and no
more data is on the receive queue. Correct behavior is to return 0 to the
user and the user can then close the socket. The EAGAIN causes many apps
to retry which masks the problem. Eventually the socket is evicted from
the sockmap because its released from sockmap sock free handling. The
issue creates a delay and can cause some errors on application side.
To fix this check on sk_msg_recvmsg side if length is zero and FIN flag
is set then set return to zero. A selftest will be added to check this
condition.
Fixes: 04919bed948dc ("tcp: Introduce tcp_read_skb()")
Tested-by: William Findlay <will@isovalent.com>
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
net/ipv4/tcp_bpf.c | 31 +++++++++++++++++++++++++++++++
1 file changed, 31 insertions(+)
diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c
index cf26d65ca389..3a0f43f3afd8 100644
--- a/net/ipv4/tcp_bpf.c
+++ b/net/ipv4/tcp_bpf.c
@@ -174,6 +174,24 @@ static int tcp_msg_wait_data(struct sock *sk, struct sk_psock *psock,
return ret;
}
+static bool is_next_msg_fin(struct sk_psock *psock)
+{
+ struct scatterlist *sge;
+ struct sk_msg *msg_rx;
+ int i;
+
+ msg_rx = sk_psock_peek_msg(psock);
+ i = msg_rx->sg.start;
+ sge = sk_msg_elem(msg_rx, i);
+ if (!sge->length) {
+ struct sk_buff *skb = msg_rx->skb;
+
+ if (skb && TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
+ return true;
+ }
+ return false;
+}
+
static int tcp_bpf_recvmsg_parser(struct sock *sk,
struct msghdr *msg,
size_t len,
@@ -193,6 +211,19 @@ static int tcp_bpf_recvmsg_parser(struct sock *sk,
lock_sock(sk);
msg_bytes_ready:
copied = sk_msg_recvmsg(sk, psock, msg, len, flags);
+ /* The typical case for EFAULT is the socket was gracefully
+ * shutdown with a FIN pkt. So check here the other case is
+ * some error on copy_page_to_iter which would be unexpected.
+ * On fin return correct return code to zero.
+ */
+ if (copied == -EFAULT) {
+ bool is_fin = is_next_msg_fin(psock);
+
+ if (is_fin) {
+ copied = 0;
+ goto out;
+ }
+ }
if (!copied) {
long timeo;
int data;
--
2.33.0
next prev parent reply other threads:[~2023-03-27 17:55 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-27 17:54 [PATCH bpf v2 00/11] bpf sockmap fixes John Fastabend
2023-03-27 17:54 ` [PATCH bpf v2 01/12] bpf: sockmap, pass skb ownership through read_skb John Fastabend
2023-03-28 10:42 ` Jakub Sitnicki
2023-03-27 17:54 ` [PATCH bpf v2 02/12] bpf: sockmap, convert schedule_work into delayed_work John Fastabend
2023-03-28 12:09 ` Jakub Sitnicki
2023-03-28 21:56 ` John Fastabend
2023-03-29 11:09 ` Jakub Sitnicki
2023-03-27 17:54 ` [PATCH bpf v2 03/12] bpf: sockmap, improved check for empty queue John Fastabend
2023-03-29 12:24 ` Jakub Sitnicki
2023-04-01 0:59 ` John Fastabend
2023-04-03 8:42 ` Jakub Sitnicki
2023-03-27 17:54 ` John Fastabend [this message]
2023-04-03 11:11 ` [PATCH bpf v2 04/12] bpf: sockmap, handle fin correctly Jakub Sitnicki
2023-04-03 21:05 ` John Fastabend
2023-04-04 10:11 ` Jakub Sitnicki
2023-03-27 17:54 ` [PATCH bpf v2 05/12] bpf: sockmap, TCP data stall on recv before accept John Fastabend
2023-03-27 17:54 ` [PATCH bpf v2 06/12] bpf: sockmap, wake up polling after data copy John Fastabend
2023-03-27 17:54 ` [PATCH bpf v2 07/12] bpf: sockmap incorrectly handling copied_seq John Fastabend
2023-03-27 17:54 ` [PATCH bpf v2 08/12] bpf: sockmap, pull socket helpers out of listen test for general use John Fastabend
2023-03-27 17:54 ` [PATCH bpf v2 09/12] bpf: sockmap, build helper to create connected socket pair John Fastabend
2023-03-27 17:54 ` [PATCH bpf v2 10/12] bpf: sockmap, test shutdown() correctly exits epoll and recv()=0 John Fastabend
2023-03-27 17:54 ` [PATCH bpf v2 11/12] bpf: sockmap, test FIONREAD returns correct bytes in rx buffer John Fastabend
2023-03-27 17:54 ` [PATCH bpf v2 12/12] bpf: sockmap, test FIONREAD returns correct bytes in rx buffer with drops John Fastabend
2023-03-27 18:17 ` [PATCH bpf v2 00/11] bpf sockmap fixes 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=20230327175446.98151-5-john.fastabend@gmail.com \
--to=john.fastabend@gmail.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=cong.wang@bytedance.com \
--cc=daniel@iogearbox.net \
--cc=edumazet@google.com \
--cc=jakub@cloudflare.com \
--cc=lmb@isovalent.com \
--cc=netdev@vger.kernel.org \
--cc=will@isovalent.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).