From: John Fastabend <john.fastabend@gmail.com>
To: ast@kernel.org, daniel@iogearbox.net, jakub@cloudflare.com
Cc: john.fastabend@gmail.com, bpf@vger.kernel.org, netdev@vger.kernel.org
Subject: [bpf PATCH v2 2/6] bpf, sockmap: Ensure SO_RCVBUF memory is observed on ingress redirect
Date: Thu, 12 Nov 2020 15:27:01 -0800 [thread overview]
Message-ID: <160522362100.135009.18395216656832785566.stgit@john-XPS-13-9370> (raw)
In-Reply-To: <160522352433.135009.15329422887113794062.stgit@john-XPS-13-9370>
Fix sockmap sk_skb programs so that they observe sk_rcvbuf limits. This
allows users to tune SO_RCVBUF and sockmap will honor them.
We can refactor the if(charge) case out in later patches. But, keep this
fix to the point.
Fixes: 51199405f9672 ("bpf: skb_verdict, support SK_PASS on RX BPF path")
Suggested-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
net/core/skmsg.c | 20 ++++++++++++++++----
net/ipv4/tcp_bpf.c | 3 ++-
2 files changed, 18 insertions(+), 5 deletions(-)
diff --git a/net/core/skmsg.c b/net/core/skmsg.c
index 654182ecf87b..fe44280c033e 100644
--- a/net/core/skmsg.c
+++ b/net/core/skmsg.c
@@ -170,10 +170,12 @@ static int sk_msg_free_elem(struct sock *sk, struct sk_msg *msg, u32 i,
struct scatterlist *sge = sk_msg_elem(msg, i);
u32 len = sge->length;
- if (charge)
- sk_mem_uncharge(sk, len);
- if (!msg->skb)
+ /* When the skb owns the memory we free it from consume_skb path. */
+ if (!msg->skb) {
+ if (charge)
+ sk_mem_uncharge(sk, len);
put_page(sg_page(sge));
+ }
memset(sge, 0, sizeof(*sge));
return len;
}
@@ -403,6 +405,9 @@ static int sk_psock_skb_ingress(struct sk_psock *psock, struct sk_buff *skb)
int copied = 0, num_sge;
struct sk_msg *msg;
+ if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf)
+ return -EAGAIN;
+
msg = kzalloc(sizeof(*msg), __GFP_NOWARN | GFP_ATOMIC);
if (unlikely(!msg))
return -EAGAIN;
@@ -418,7 +423,14 @@ static int sk_psock_skb_ingress(struct sk_psock *psock, struct sk_buff *skb)
return num_sge;
}
- sk_mem_charge(sk, skb->len);
+ /* This will transition ownership of the data from the socket where
+ * the BPF program was run initiating the redirect to the socket
+ * we will eventually receive this data on. The data will be released
+ * from skb_consume found in __tcp_bpf_recvmsg() after its been copied
+ * into user buffers.
+ */
+ skb_set_owner_r(skb, sk);
+
copied = skb->len;
msg->sg.start = 0;
msg->sg.size = copied;
diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c
index 8e950b0bfabc..bc7d2a586e18 100644
--- a/net/ipv4/tcp_bpf.c
+++ b/net/ipv4/tcp_bpf.c
@@ -45,7 +45,8 @@ int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock,
if (likely(!peek)) {
sge->offset += copy;
sge->length -= copy;
- sk_mem_uncharge(sk, copy);
+ if (!msg_rx->skb)
+ sk_mem_uncharge(sk, copy);
msg_rx->sg.size -= copy;
if (!sge->length) {
next prev parent reply other threads:[~2020-11-12 23:27 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-12 23:26 [bpf PATCH v2 0/6] sockmap fixes John Fastabend
2020-11-12 23:26 ` [bpf PATCH v2 1/6] bpf, sockmap: fix partial copy_page_to_iter so progress can still be made John Fastabend
2020-11-12 23:27 ` John Fastabend [this message]
2020-11-12 23:27 ` [bpf PATCH v2 3/6] bpf, sockmap: Use truesize with sk_rmem_schedule() John Fastabend
2020-11-12 23:27 ` [bpf PATCH v2 4/6] bpf, sockmap: Avoid returning unneeded EAGAIN when redirecting to self John Fastabend
2020-11-12 23:27 ` [bpf PATCH v2 5/6] bpf, sockmap: Handle memory acct if skb_verdict prog redirects " John Fastabend
2020-11-16 14:31 ` Jakub Sitnicki
2020-11-16 22:28 ` John Fastabend
2020-11-12 23:28 ` [bpf PATCH v2 6/6] bpf, sockmap: Avoid failures from skb_to_sgvec when skb has frag_list John Fastabend
2020-11-16 14:49 ` [bpf PATCH v2 0/6] sockmap fixes Jakub Sitnicki
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=160522362100.135009.18395216656832785566.stgit@john-XPS-13-9370 \
--to=john.fastabend@gmail.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=jakub@cloudflare.com \
--cc=netdev@vger.kernel.org \
/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).