From: Geliang Tang <tanggeliang@kylinos.cn>
It's necessary to traverse all subflows on the conn_list of an MPTCP
socket and then call kfunc to modify the fields of each subflow. In
kernel space, mptcp_for_each_subflow() helper is used for this:
mptcp_for_each_subflow(msk, subflow)
kfunc(subflow);
But in the MPTCP BPF program, this has not yet been implemented. As
Martin suggested recently, this conn_list walking + modify-by-kfunc
usage fits the bpf_iter use case. So this patch adds a new bpf_iter
type named "mptcp_subflow" to do this and implements its helpers
bpf_iter_mptcp_subflow_new()/_next()/_destroy().
Then bpf_for_each() for mptcp_subflow can be used in BPF program like
this:
bpf_rcu_read_lock();
bpf_for_each(mptcp_subflow, subflow, msk)
kfunc(subflow);
bpf_rcu_read_unlock();
Suggested-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
net/mptcp/bpf.c | 51 ++++++++++++++++++++++++++++++++++++++++++++
net/mptcp/protocol.h | 6 ++++++
2 files changed, 57 insertions(+)
diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
index 9672a70c24b0..799264119891 100644
--- a/net/mptcp/bpf.c
+++ b/net/mptcp/bpf.c
@@ -204,10 +204,59 @@ static const struct btf_kfunc_id_set bpf_mptcp_fmodret_set = {
.set = &bpf_mptcp_fmodret_ids,
};
+struct bpf_iter__mptcp_subflow {
+ __bpf_md_ptr(struct bpf_iter_meta *, meta);
+ __bpf_md_ptr(struct mptcp_sock *, msk);
+ __bpf_md_ptr(struct list_head *, pos);
+};
+
+struct bpf_iter_mptcp_subflow {
+ __u64 __opaque[2];
+} __attribute__((aligned(8)));
+
+struct bpf_iter_mptcp_subflow_kern {
+ struct mptcp_sock *msk;
+ struct list_head *pos;
+} __attribute__((aligned(8)));
+
__diag_push();
__diag_ignore_all("-Wmissing-prototypes",
"kfuncs which will be used in BPF programs");
+__bpf_kfunc_start_defs();
+
+__bpf_kfunc int bpf_iter_mptcp_subflow_new(struct bpf_iter_mptcp_subflow *it,
+ struct mptcp_sock *msk)
+{
+ struct bpf_iter_mptcp_subflow_kern *kit = (void *)it;
+
+ if (!msk)
+ return -EINVAL;
+
+ kit->msk = msk;
+ kit->pos = &msk->conn_list;
+ return 0;
+}
+
+__bpf_kfunc struct mptcp_subflow_context *
+bpf_iter_mptcp_subflow_next(struct bpf_iter_mptcp_subflow *it)
+{
+ struct bpf_iter_mptcp_subflow_kern *kit = (void *)it;
+ struct mptcp_subflow_context *subflow;
+ struct mptcp_sock *msk = kit->msk;
+
+ subflow = list_entry((kit->pos)->next, struct mptcp_subflow_context, node);
+ if (!msk || list_entry_is_head(subflow, &msk->conn_list, node))
+ return NULL;
+
+ kit->pos = &subflow->node;
+ return subflow;
+}
+
+__bpf_kfunc void bpf_iter_mptcp_subflow_destroy(struct bpf_iter_mptcp_subflow *it)
+{
+}
+
__bpf_kfunc struct mptcp_subflow_context *
bpf_mptcp_subflow_ctx_by_pos(const struct mptcp_sched_data *data, unsigned int pos)
{
@@ -221,6 +270,8 @@ __bpf_kfunc bool bpf_mptcp_subflow_queues_empty(struct sock *sk)
return tcp_rtx_queue_empty(sk);
}
+__bpf_kfunc_end_defs();
+
__diag_pop();
BTF_KFUNCS_START(bpf_mptcp_sched_kfunc_ids)
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index d25d2dac88a5..b3f5254e3c0d 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -715,6 +715,12 @@ void mptcp_subflow_queue_clean(struct sock *sk, struct sock *ssk);
void mptcp_sock_graft(struct sock *sk, struct socket *parent);
u64 mptcp_wnd_end(const struct mptcp_sock *msk);
void mptcp_set_timeout(struct sock *sk);
+struct bpf_iter_mptcp_subflow;
+int bpf_iter_mptcp_subflow_new(struct bpf_iter_mptcp_subflow *it,
+ struct mptcp_sock *msk);
+struct mptcp_subflow_context *
+bpf_iter_mptcp_subflow_next(struct bpf_iter_mptcp_subflow *it);
+void bpf_iter_mptcp_subflow_destroy(struct bpf_iter_mptcp_subflow *it);
bool bpf_mptcp_subflow_queues_empty(struct sock *sk);
struct mptcp_subflow_context *
bpf_mptcp_subflow_ctx_by_pos(const struct mptcp_sched_data *data, unsigned int pos);
--
2.43.0On Mon, 2024-09-09 at 09:36 +0800, Geliang Tang wrote:
> From: Geliang Tang <tanggeliang@kylinos.cn>
>
> It's necessary to traverse all subflows on the conn_list of an MPTCP
> socket and then call kfunc to modify the fields of each subflow. In
> kernel space, mptcp_for_each_subflow() helper is used for this:
>
> mptcp_for_each_subflow(msk, subflow)
> kfunc(subflow);
>
> But in the MPTCP BPF program, this has not yet been implemented. As
> Martin suggested recently, this conn_list walking + modify-by-kfunc
> usage fits the bpf_iter use case. So this patch adds a new bpf_iter
> type named "mptcp_subflow" to do this and implements its helpers
> bpf_iter_mptcp_subflow_new()/_next()/_destroy().
>
> Then bpf_for_each() for mptcp_subflow can be used in BPF program like
> this:
>
> bpf_rcu_read_lock();
> bpf_for_each(mptcp_subflow, subflow, msk)
> kfunc(subflow);
> bpf_rcu_read_unlock();
>
> Suggested-by: Martin KaFai Lau <martin.lau@kernel.org>
> Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
> ---
> net/mptcp/bpf.c | 51
> ++++++++++++++++++++++++++++++++++++++++++++
> net/mptcp/protocol.h | 6 ++++++
> 2 files changed, 57 insertions(+)
>
> diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
> index 9672a70c24b0..799264119891 100644
> --- a/net/mptcp/bpf.c
> +++ b/net/mptcp/bpf.c
> @@ -204,10 +204,59 @@ static const struct btf_kfunc_id_set
> bpf_mptcp_fmodret_set = {
> .set = &bpf_mptcp_fmodret_ids,
> };
>
> +struct bpf_iter__mptcp_subflow {
> + __bpf_md_ptr(struct bpf_iter_meta *, meta);
> + __bpf_md_ptr(struct mptcp_sock *, msk);
> + __bpf_md_ptr(struct list_head *, pos);
> +};
This bpf_iter__mptcp_subflow struct should be dropped too.
> +
> +struct bpf_iter_mptcp_subflow {
> + __u64 __opaque[2];
> +} __attribute__((aligned(8)));
> +
> +struct bpf_iter_mptcp_subflow_kern {
> + struct mptcp_sock *msk;
> + struct list_head *pos;
> +} __attribute__((aligned(8)));
> +
> __diag_push();
> __diag_ignore_all("-Wmissing-prototypes",
> "kfuncs which will be used in BPF programs");
Duplicate with __bpf_kfunc_start_defs/__bpf_kfunc_end_defs,
__diag_push, __diag_pop and __diag_ignore_all should be dropped.
>
> +__bpf_kfunc_start_defs();
> +
> +__bpf_kfunc int bpf_iter_mptcp_subflow_new(struct
> bpf_iter_mptcp_subflow *it,
> + struct mptcp_sock *msk)
> +{
> + struct bpf_iter_mptcp_subflow_kern *kit = (void *)it;
> +
> + if (!msk)
> + return -EINVAL;
> +
> + kit->msk = msk;
> + kit->pos = &msk->conn_list;
> + return 0;
> +}
> +
> +__bpf_kfunc struct mptcp_subflow_context *
> +bpf_iter_mptcp_subflow_next(struct bpf_iter_mptcp_subflow *it)
> +{
> + struct bpf_iter_mptcp_subflow_kern *kit = (void *)it;
> + struct mptcp_subflow_context *subflow;
> + struct mptcp_sock *msk = kit->msk;
> +
> + subflow = list_entry((kit->pos)->next, struct
> mptcp_subflow_context, node);
> + if (!msk || list_entry_is_head(subflow, &msk->conn_list,
> node))
> + return NULL;
> +
> + kit->pos = &subflow->node;
> + return subflow;
> +}
> +
> +__bpf_kfunc void bpf_iter_mptcp_subflow_destroy(struct
> bpf_iter_mptcp_subflow *it)
> +{
> +}
> +
> __bpf_kfunc struct mptcp_subflow_context *
> bpf_mptcp_subflow_ctx_by_pos(const struct mptcp_sched_data *data,
> unsigned int pos)
> {
> @@ -221,6 +270,8 @@ __bpf_kfunc bool
> bpf_mptcp_subflow_queues_empty(struct sock *sk)
> return tcp_rtx_queue_empty(sk);
> }
>
> +__bpf_kfunc_end_defs();
> +
> __diag_pop();
>
> BTF_KFUNCS_START(bpf_mptcp_sched_kfunc_ids)
> diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
> index d25d2dac88a5..b3f5254e3c0d 100644
> --- a/net/mptcp/protocol.h
> +++ b/net/mptcp/protocol.h
> @@ -715,6 +715,12 @@ void mptcp_subflow_queue_clean(struct sock *sk,
> struct sock *ssk);
> void mptcp_sock_graft(struct sock *sk, struct socket *parent);
> u64 mptcp_wnd_end(const struct mptcp_sock *msk);
> void mptcp_set_timeout(struct sock *sk);
> +struct bpf_iter_mptcp_subflow;
> +int bpf_iter_mptcp_subflow_new(struct bpf_iter_mptcp_subflow *it,
> + struct mptcp_sock *msk);
> +struct mptcp_subflow_context *
> +bpf_iter_mptcp_subflow_next(struct bpf_iter_mptcp_subflow *it);
> +void bpf_iter_mptcp_subflow_destroy(struct bpf_iter_mptcp_subflow
> *it);
No need to add these declarations, since "-Wmissing-declarations" is
ignored in __bpf_kfunc_start_defs.
Will update in v3.
Thanks,
-Geliang
> bool bpf_mptcp_subflow_queues_empty(struct sock *sk);
> struct mptcp_subflow_context *
> bpf_mptcp_subflow_ctx_by_pos(const struct mptcp_sched_data *data,
> unsigned int pos);
© 2016 - 2026 Red Hat, Inc.