-
Notifications
You must be signed in to change notification settings - Fork 109
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
tcp_bpf: update the rmem scheduling for ingress redirection #7904
base: bpf_base
Are you sure you want to change the base?
Conversation
Upstream branch: 2aa587f |
086ca61
to
0e95e1e
Compare
Upstream branch: 19039f2 |
e826cfe
to
8e86945
Compare
0e95e1e
to
2a4bf0e
Compare
Upstream branch: 98cd619 |
8e86945
to
47c4d14
Compare
2a4bf0e
to
b3e861b
Compare
Upstream branch: db123e4 |
47c4d14
to
b90a2a7
Compare
b3e861b
to
5dede3a
Compare
Upstream branch: 5ac9b4e |
b90a2a7
to
1640c8b
Compare
5dede3a
to
6d81e01
Compare
Upstream branch: 5ac9b4e |
When bpf_tcp_ingress() is called, the skmsg is being redirected to the ingress of the destination socket. Therefore, we should charge its receive socket buffer, instead of sending socket buffer. Because sk_rmem_schedule() tests pfmemalloc of skb, we need to introduce a wrapper and call it for skmsg. Signed-off-by: Cong Wang <cong.wang@bytedance.com>
Although we sk_rmem_schedule and add sk_msg to the ingress_msg of sk_redir in bpf_tcp_ingress, we do not update sk_rmem_alloc. As a result, except for the global memory limit, the rmem of sk_redir is nearly unlimited. Thus, add sk_rmem_alloc related logic to limit the recv buffer. Signed-off-by: Zijian Zhang <zijianzhang@bytedance.com>
1640c8b
to
0088ea5
Compare
Pull request for series with
subject: tcp_bpf: update the rmem scheduling for ingress redirection
version: 1
url: https://patchwork.kernel.org/project/netdevbpf/list/?series=900001