Series comparison

-[PATCH mptcp-next v6 00/10] use bpf_iter in bpf schedulers
+[PATCH mptcp-next v13 0/9] use bpf_iter in bpf schedulers
 From: Geliang Tang <tanggeliang@kylinos.cn>
+v13:
+ - use '__ign' suffix to ignore the argument type checks of
+   bpf_mptcp_subflow_ctx() and bpf_sk_stream_memory_free(),
+   instead of adding a new helper bpf_mptcp_send_info_to_ssk().
+ - use 'bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk)' instead
+   of using 'bpf_for_each(mptcp_subflow, subflow, msk)'.
+ - keep struct mptcp_sched_data for future use.
+Depends on:
+ - Squash to "Add mptcp_subflow bpf_iter support", v2
+Based-on: <cover.1738470660.git.tanggeliang@kylinos.cn>
+v12:
+ - drop struct mptcp_sched_data.
+ - rebased on "split get_subflow interface into two" v2.
+v11:
+If another squash-to patchset (Squash to "Add mptcp_subflow bpf_iter
+support") under review is merged before this set, v10 will fail to run.
+v11 fixes this issue and can run regardless of whether it is merged
+before or after the squash-to patchset.
+Compared with v10, only patches 3, 5, and 8 have been modified:
+ - use mptcp_subflow_tcp_sock instead of bpf_mptcp_subflow_tcp_sock in
+   patch 3 and patch 5.
+ - drop bpf_mptcp_sched_kfunc_set, use bpf_mptcp_common_kfunc_set instead
+   in patch 8.
+v10:
+ - drop mptcp_subflow_set_scheduled() helper and WRITE_ONCE() in BPF.
+ - add new bpf helper bpf_mptcp_send_info_to_ssk() for burst scheduler.
+v9:
+ - merge 'Fixes for "use bpf_iter in bpf schedulers" v8' into this set.
+ - rebased on "add netns helpers" v4
+v8:
+ - address Mat's comments in v7.
+ - move sk_stream_memory_free check inside bpf_for_each() loop.
+ - implement mptcp_subflow_set_scheduled helper in BPF.
+ - add cleanup patches into this set again.
+v7:
+ - move cleanup patches out of this set.
+ - rebased.
 v6:
  - rebased to "add mptcp_subflow bpf_iter" v10
 v5:
 ...
 With the newly added mptcp_subflow bpf_iter, we can get rid of the
 subflows array "contexts" in struct mptcp_sched_data. This set
 uses bpf_for_each(mptcp_subflow) helper to update all the bpf
 schedules:
-        bpf_for_each(mptcp_subflow, subflow, msk) {
+        bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk) {
                 ... ...
                 mptcp_subflow_set_scheduled(subflow, true);
         }
-Depends on:
+Geliang Tang (9):
- - "add mptcp_subflow bpf_iter" v10
+  Squash to "bpf: Register mptcp common kfunc set"
 Based-on: <cover.1729063444.git.tanggeliang@kylinos.cn>
 Geliang Tang (10):
   Revert "mptcp: add sched_data helpers"
-  Squash to "bpf: Add bpf_mptcp_sched_ops"
+  Squash to "bpf: Export mptcp packet scheduler helpers"
   Squash to "bpf: Add bpf_mptcp_sched_kfunc_set"
   Squash to "selftests/bpf: Add bpf_first scheduler & test"
   Squash to "selftests/bpf: Add bpf_bkup scheduler & test"
   Squash to "selftests/bpf: Add bpf_rr scheduler & test"
   Squash to "selftests/bpf: Add bpf_red scheduler & test"
   Squash to "selftests/bpf: Add bpf_burst scheduler & test"
   mptcp: drop subflow contexts in mptcp_sched_data
-  Squash to "selftests/bpf: Add bpf scheduler test" - drop
-    has_bytes_sent
- include/net/mptcp.h                           |  2 -
+ include/net/mptcp.h                           |  3 -
- net/mptcp/bpf.c                               | 24 ++----
+ net/mptcp/bpf.c                               | 52 +++++++------
  net/mptcp/protocol.h                          |  2 -
  net/mptcp/sched.c                             | 22 ------
- .../testing/selftests/bpf/prog_tests/mptcp.c  | 48 ++++++------
  tools/testing/selftests/bpf/progs/mptcp_bpf.h |  3 -
  .../selftests/bpf/progs/mptcp_bpf_bkup.c      | 16 +---
- .../selftests/bpf/progs/mptcp_bpf_burst.c     | 77 ++++++++++---------
+ .../selftests/bpf/progs/mptcp_bpf_burst.c     | 78 +++++++------------
  .../selftests/bpf/progs/mptcp_bpf_bytes.c     | 39 ++++++++++
  .../selftests/bpf/progs/mptcp_bpf_first.c     |  8 +-
  .../selftests/bpf/progs/mptcp_bpf_red.c       |  8 +-
- .../selftests/bpf/progs/mptcp_bpf_rr.c        | 24 +++---
+ .../selftests/bpf/progs/mptcp_bpf_rr.c        | 31 ++++----
-files changed, 133 insertions(+), 140 deletions(-)
+files changed, 80 insertions(+), 143 deletions(-)
  create mode 100644 tools/testing/selftests/bpf/progs/mptcp_bpf_bytes.c
 --
 .43.0

-[PATCH mptcp-next v6 02/10] Squash to "bpf: Add bpf_mptcp_sched_ops"
+[PATCH mptcp-next v13 1/9] Squash to "bpf: Register mptcp common kfunc set"
 From: Geliang Tang <tanggeliang@kylinos.cn>
-Please update the subject to
+Instead of adding a new BPF function bpf_mptcp_send_info_to_ssk() in
+v12, this patch uses a much more simpler approach, which using '__ign'
-    bpf: Add mptcp packet scheduler struct_ops
+suffix for the argument of bpf_mptcp_subflow_ctx() to let BPF to
+ignore the type check of this argument.
 Drop mptcp_sock_type and mptcp_subflow_type.
 Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
 ---
- net/mptcp/bpf.c | 15 ++++++---------
+ net/mptcp/bpf.c | 8 ++++----
-file changed, 6 insertions(+), 9 deletions(-)
+file changed, 4 insertions(+), 4 deletions(-)
 diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/mptcp/bpf.c
 +++ b/net/mptcp/bpf.c
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ struct bpf_iter_mptcp_subflow_kern {
+ __bpf_kfunc_start_defs();
- #ifdef CONFIG_BPF_JIT
- static struct bpf_struct_ops bpf_mptcp_sched_ops;
+ __bpf_kfunc static struct mptcp_subflow_context *
--static const struct btf_type *mptcp_sock_type, *mptcp_subflow_type __read_mostly;
+-bpf_mptcp_subflow_ctx(const struct sock *sk)
- static u32 mptcp_sock_id, mptcp_subflow_id;
++bpf_mptcp_subflow_ctx(const struct sock *sk__ign)
 +/* MPTCP BPF packet scheduler */
 +
  static const struct bpf_func_proto *
  bpf_mptcp_sched_get_func_proto(enum bpf_func_id func_id,
                     const struct bpf_prog *prog)
@@ -XXX,XX +XXX,XX @@ static int bpf_mptcp_sched_btf_struct_access(struct bpf_verifier_log *log,
                           const struct bpf_reg_state *reg,
                           int off, int size)
  {
--    const struct btf_type *t;
+-    if (sk && sk_fullsock(sk) &&
-+    u32 id = reg->btf_id;
+-        sk->sk_protocol == IPPROTO_TCP && sk_is_mptcp(sk))
-     size_t end;
+-        return mptcp_subflow_ctx(sk);
++    if (sk__ign && sk_fullsock(sk__ign) &&
--    t = btf_type_by_id(reg->btf, reg->btf_id);
++        sk__ign->sk_protocol == IPPROTO_TCP && sk_is_mptcp(sk__ign))
--
++        return mptcp_subflow_ctx(sk__ign);
--    if (t == mptcp_sock_type) {
-+    if (id == mptcp_sock_id) {
+     return NULL;
          switch (off) {
          case offsetof(struct mptcp_sock, snd_burst):
              end = offsetofend(struct mptcp_sock, snd_burst);
@@ -XXX,XX +XXX,XX @@ static int bpf_mptcp_sched_btf_struct_access(struct bpf_verifier_log *log,
                  off);
              return -EACCES;
          }
 -    } else if (t == mptcp_subflow_type) {
 +    } else if (id == mptcp_subflow_id) {
          switch (off) {
          case offsetof(struct mptcp_subflow_context, avg_pacing_rate):
              end = offsetofend(struct mptcp_subflow_context, avg_pacing_rate);
@@ -XXX,XX +XXX,XX @@ static int bpf_mptcp_sched_btf_struct_access(struct bpf_verifier_log *log,
      if (off + size > end) {
          bpf_log(log, "access beyond %s at off %u size %u ended at %zu",
 -            t == mptcp_sock_type ? "mptcp_sock" : "mptcp_subflow_context",
 +            id == mptcp_sock_id ? "mptcp_sock" : "mptcp_subflow_context",
              off, size, end);
          return -EACCES;
      }
@@ -XXX,XX +XXX,XX @@ static int bpf_mptcp_sched_init(struct btf *btf)
      if (type_id < 0)
          return -EINVAL;
      mptcp_sock_id = type_id;
 -    mptcp_sock_type = btf_type_by_id(btf, mptcp_sock_id);
      type_id = btf_find_by_name_kind(btf, "mptcp_subflow_context",
                      BTF_KIND_STRUCT);
      if (type_id < 0)
          return -EINVAL;
      mptcp_subflow_id = type_id;
 -    mptcp_subflow_type = btf_type_by_id(btf, mptcp_subflow_id);
      return 0;
  }
 --
 .43.0

-[PATCH mptcp-next v6 01/10] Revert "mptcp: add sched_data helpers"
+[PATCH mptcp-next v13 2/9] Revert "mptcp: add sched_data helpers"
 From: Geliang Tang <tanggeliang@kylinos.cn>
-Drop this patch.
+Drop this patch. bpf_mptcp_subflow_ctx_by_pos and
 mptcp_sched_data_set_contexts are uesless now.
 Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
 ---
  net/mptcp/bpf.c      |  8 --------
  net/mptcp/protocol.h |  2 --
 ...
 diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/mptcp/bpf.c
 +++ b/net/mptcp/bpf.c
-@@ -XXX,XX +XXX,XX @@ __bpf_kfunc static void bpf_mptcp_sock_release(struct mptcp_sock *msk)
+@@ -XXX,XX +XXX,XX @@ bpf_iter_mptcp_subflow_destroy(struct bpf_iter_mptcp_subflow *it)
-     WARN_ON_ONCE(!sk || !refcount_dec_not_one(&sk->sk_refcnt));
+ {
  }
 -__bpf_kfunc struct mptcp_subflow_context *
 -bpf_mptcp_subflow_ctx_by_pos(const struct mptcp_sched_data *data, unsigned int pos)
 -{
 ...
 -
  int mptcp_sched_get_send(struct mptcp_sock *msk)
  {
      struct mptcp_subflow_context *subflow;
 @@ -XXX,XX +XXX,XX @@ int mptcp_sched_get_send(struct mptcp_sock *msk)
-     data.reinject = false;
      if (msk->sched == &mptcp_sched_default || !msk->sched)
-         return mptcp_sched_default_get_subflow(msk, &data);
+         return mptcp_sched_default_get_send(msk, &data);
 -    mptcp_sched_data_set_contexts(msk, &data);
-     return msk->sched->get_subflow(msk, &data);
+     return msk->sched->get_send(msk, &data);
  }
 @@ -XXX,XX +XXX,XX @@ int mptcp_sched_get_retrans(struct mptcp_sock *msk)
-     data.reinject = true;
      if (msk->sched == &mptcp_sched_default || !msk->sched)
-         return mptcp_sched_default_get_subflow(msk, &data);
+         return mptcp_sched_default_get_retrans(msk, &data);
 -    mptcp_sched_data_set_contexts(msk, &data);
-     return msk->sched->get_subflow(msk, &data);
+     if (msk->sched->get_retrans)
- }
+         return msk->sched->get_retrans(msk, &data);
      return msk->sched->get_send(msk, &data);
 --
 .43.0

-[PATCH mptcp-next v6 03/10] Squash to "bpf: Add bpf_mptcp_sched_kfunc_set"
+[PATCH mptcp-next v13 3/9] Squash to "bpf: Export mptcp packet scheduler helpers"
 From: Geliang Tang <tanggeliang@kylinos.cn>
-Please update the subject to
-    "bpf: Export mptcp packet scheduler helpers"
 Remove bpf_mptcp_subflow_ctx_by_pos from BPF kfunc set.
+Drop bpf_mptcp_sched_kfunc_set, use bpf_mptcp_common_kfunc_set instead.
+Add new helpers bpf_mptcp_subflow_tcp_sock() and
+bpf_sk_stream_memory_free().
 Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
 ---
- net/mptcp/bpf.c | 1 -
+ net/mptcp/bpf.c | 38 ++++++++++++++++++++++++--------------
-file changed, 1 deletion(-)
+file changed, 24 insertions(+), 14 deletions(-)
 diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/mptcp/bpf.c
 +++ b/net/mptcp/bpf.c
-@@ -XXX,XX +XXX,XX @@ static const struct btf_kfunc_id_set bpf_mptcp_common_kfunc_set = {
+@@ -XXX,XX +XXX,XX @@ bpf_mptcp_subflow_ctx(const struct sock *sk__ign)
+     return NULL;
- BTF_KFUNCS_START(bpf_mptcp_sched_kfunc_ids)
+ }
 +__bpf_kfunc static struct sock *
 +bpf_mptcp_subflow_tcp_sock(const struct mptcp_subflow_context *subflow)
 +{
 +    if (!subflow)
 +        return NULL;
 +
 +    return mptcp_subflow_tcp_sock(subflow);
 +}
 +
  __bpf_kfunc static int
  bpf_iter_mptcp_subflow_new(struct bpf_iter_mptcp_subflow *it,
                 struct sock *sk)
@@ -XXX,XX +XXX,XX @@ __bpf_kfunc static bool bpf_mptcp_subflow_queues_empty(struct sock *sk)
      return tcp_rtx_queue_empty(sk);
  }
 +__bpf_kfunc static bool bpf_sk_stream_memory_free(const struct sock *sk__ign)
 +{
 +    if (sk__ign && sk_fullsock(sk__ign) &&
 +        sk__ign->sk_protocol == IPPROTO_TCP && sk_is_mptcp(sk__ign))
 +        return sk_stream_memory_free(sk__ign);
 +
 +    return NULL;
 +}
 +
  __bpf_kfunc_end_defs();
  BTF_KFUNCS_START(bpf_mptcp_common_kfunc_ids)
  BTF_ID_FLAGS(func, bpf_mptcp_subflow_ctx, KF_RET_NULL)
 +BTF_ID_FLAGS(func, bpf_mptcp_subflow_tcp_sock, KF_RET_NULL)
  BTF_ID_FLAGS(func, bpf_iter_mptcp_subflow_new, KF_ITER_NEW | KF_TRUSTED_ARGS)
  BTF_ID_FLAGS(func, bpf_iter_mptcp_subflow_next, KF_ITER_NEXT | KF_RET_NULL)
  BTF_ID_FLAGS(func, bpf_iter_mptcp_subflow_destroy, KF_ITER_DESTROY)
 -BTF_KFUNCS_END(bpf_mptcp_common_kfunc_ids)
 -
 -static const struct btf_kfunc_id_set bpf_mptcp_common_kfunc_set = {
 -    .owner    = THIS_MODULE,
 -    .set    = &bpf_mptcp_common_kfunc_ids,
 -};
 -
 -BTF_KFUNCS_START(bpf_mptcp_sched_kfunc_ids)
  BTF_ID_FLAGS(func, mptcp_subflow_set_scheduled)
 -BTF_ID_FLAGS(func, bpf_mptcp_subflow_ctx_by_pos)
  BTF_ID_FLAGS(func, mptcp_subflow_active)
  BTF_ID_FLAGS(func, mptcp_set_timeout)
  BTF_ID_FLAGS(func, mptcp_wnd_end)
+-BTF_ID_FLAGS(func, tcp_stream_memory_free)
++BTF_ID_FLAGS(func, bpf_sk_stream_memory_free, KF_RET_NULL)
+ BTF_ID_FLAGS(func, bpf_mptcp_subflow_queues_empty)
+ BTF_ID_FLAGS(func, mptcp_pm_subflow_chk_stale, KF_SLEEPABLE)
+-BTF_KFUNCS_END(bpf_mptcp_sched_kfunc_ids)
++BTF_KFUNCS_END(bpf_mptcp_common_kfunc_ids)
+-static const struct btf_kfunc_id_set bpf_mptcp_sched_kfunc_set = {
++static const struct btf_kfunc_id_set bpf_mptcp_common_kfunc_set = {
+     .owner    = THIS_MODULE,
+-    .set    = &bpf_mptcp_sched_kfunc_ids,
++    .set    = &bpf_mptcp_common_kfunc_ids,
+ };
+ static int __init bpf_mptcp_kfunc_init(void)
+@@ -XXX,XX +XXX,XX @@ static int __init bpf_mptcp_kfunc_init(void)
+     ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_CGROUP_SOCKOPT,
+                            &bpf_mptcp_common_kfunc_set);
+     ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS,
+-                           &bpf_mptcp_sched_kfunc_set);
++                           &bpf_mptcp_common_kfunc_set);
+ #ifdef CONFIG_BPF_JIT
+     ret = ret ?: register_bpf_struct_ops(&bpf_mptcp_sched_ops, mptcp_sched_ops);
+ #endif
 --
 .43.0

-[PATCH mptcp-next v6 04/10] Squash to "selftests/bpf: Add bpf_first scheduler & test"
+[PATCH mptcp-next v13 4/9] Squash to "selftests/bpf: Add bpf_first scheduler & test"
 From: Geliang Tang <tanggeliang@kylinos.cn>
 Use the newly added bpf_for_each() helper to walk the conn_list.
+Drop bpf_mptcp_subflow_ctx_by_pos declaration.
 Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
 ---
  tools/testing/selftests/bpf/progs/mptcp_bpf.h       | 3 ---
  tools/testing/selftests/bpf/progs/mptcp_bpf_first.c | 8 +++++++-
 ...
 diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
 index XXXXXXX..XXXXXXX 100644
 --- a/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
 +++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
 @@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
- int BPF_PROG(bpf_first_get_subflow, struct mptcp_sock *msk,
+ int BPF_PROG(bpf_first_get_send, struct mptcp_sock *msk,
           struct mptcp_sched_data *data)
  {
 -    mptcp_subflow_set_scheduled(bpf_mptcp_subflow_ctx_by_pos(data, 0), true);
 +    struct mptcp_subflow_context *subflow;
 +
-+    bpf_for_each(mptcp_subflow, subflow, msk) {
++    subflow = bpf_mptcp_subflow_ctx(msk->first);
-+        mptcp_subflow_set_scheduled(subflow, true);
++    if (!subflow)
-+        break;
++        return -1;
 +    }
 +
++    mptcp_subflow_set_scheduled(subflow, true);
      return 0;
  }
 --
 .43.0

-[PATCH mptcp-next v6 05/10] Squash to "selftests/bpf: Add bpf_bkup scheduler & test"
+[PATCH mptcp-next v13 5/9] Squash to "selftests/bpf: Add bpf_bkup scheduler & test"
 ...
 diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
 index XXXXXXX..XXXXXXX 100644
 --- a/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
 +++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
 @@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
- int BPF_PROG(bpf_bkup_get_subflow, struct mptcp_sock *msk,
+ int BPF_PROG(bpf_bkup_get_send, struct mptcp_sock *msk,
           struct mptcp_sched_data *data)
  {
 -    int nr = -1;
 -
 -    for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
 ...
 -        subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
 -        if (!subflow)
 -            break;
 +    struct mptcp_subflow_context *subflow;
-+    bpf_for_each(mptcp_subflow, subflow, msk) {
++    bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk) {
          if (!BPF_CORE_READ_BITFIELD_PROBED(subflow, backup) ||
              !BPF_CORE_READ_BITFIELD_PROBED(subflow, request_bkup)) {
 -            nr = i;
 +            mptcp_subflow_set_scheduled(subflow, true);
              break;
 ...

-[PATCH mptcp-next v6 06/10] Squash to "selftests/bpf: Add bpf_rr scheduler & test"
+[PATCH mptcp-next v13 6/9] Squash to "selftests/bpf: Add bpf_rr scheduler & test"
 From: Geliang Tang <tanggeliang@kylinos.cn>
 Use the newly added bpf_for_each() helper to walk the conn_list.
 Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
 ---
- .../selftests/bpf/progs/mptcp_bpf_rr.c        | 24 ++++++++-----------
+ .../selftests/bpf/progs/mptcp_bpf_rr.c        | 31 +++++++++----------
-file changed, 10 insertions(+), 14 deletions(-)
+file changed, 14 insertions(+), 17 deletions(-)
 diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
 index XXXXXXX..XXXXXXX 100644
 --- a/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
 +++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
 @@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
- int BPF_PROG(bpf_rr_get_subflow, struct mptcp_sock *msk,
+ int BPF_PROG(bpf_rr_get_send, struct mptcp_sock *msk,
           struct mptcp_sched_data *data)
  {
 -    struct mptcp_subflow_context *subflow;
 +    struct mptcp_subflow_context *subflow, *next;
      struct mptcp_rr_storage *ptr;
-     struct sock *last_snd = NULL;
+-    struct sock *last_snd = NULL;
 -    int nr = 0;
      ptr = bpf_sk_storage_get(&mptcp_rr_map, msk, 0,
                   BPF_LOCAL_STORAGE_GET_F_CREATE);
-@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_rr_get_subflow, struct mptcp_sock *msk,
+     if (!ptr)
          return -1;
-     last_snd = ptr->last_snd;
+-    last_snd = ptr->last_snd;
 +    next = bpf_mptcp_subflow_ctx(msk->first);
++    if (!next)
++        return -1;
 -    for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
 -        subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
 -        if (!last_snd || !subflow)
-+    bpf_for_each(mptcp_subflow, subflow, msk) {
+-            break;
-+        if (!last_snd)
++    if (!ptr->last_snd)
-             break;
++        goto out;
 -        if (mptcp_subflow_tcp_sock(subflow) == last_snd) {
 -            if (i + 1 == MPTCP_SUBFLOWS_MAX ||
 -                !bpf_mptcp_subflow_ctx_by_pos(data, i + 1))
-+        if (bpf_mptcp_subflow_tcp_sock(subflow) == last_snd) {
++    bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk) {
 +        if (mptcp_subflow_tcp_sock(subflow) == ptr->last_snd) {
 +            subflow = bpf_iter_mptcp_subflow_next(&___it);
 +            if (!subflow)
                  break;
 -            nr = i + 1;
 ...
 -    subflow = bpf_mptcp_subflow_ctx_by_pos(data, nr);
 -    if (!subflow)
 -        return -1;
 -    mptcp_subflow_set_scheduled(subflow, true);
 -    ptr->last_snd = mptcp_subflow_tcp_sock(subflow);
++out:
 +    mptcp_subflow_set_scheduled(next, true);
-+    ptr->last_snd = bpf_mptcp_subflow_tcp_sock(next);
++    ptr->last_snd = mptcp_subflow_tcp_sock(next);
      return 0;
  }
 --
 .43.0

-[PATCH mptcp-next v6 07/10] Squash to "selftests/bpf: Add bpf_red scheduler & test"
+[PATCH mptcp-next v13 7/9] Squash to "selftests/bpf: Add bpf_red scheduler & test"
 ...
 diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
 index XXXXXXX..XXXXXXX 100644
 --- a/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
 +++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
 @@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
- int BPF_PROG(bpf_red_get_subflow, struct mptcp_sock *msk,
+ int BPF_PROG(bpf_red_get_send, struct mptcp_sock *msk,
           struct mptcp_sched_data *data)
  {
 -    for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
 -        if (!bpf_mptcp_subflow_ctx_by_pos(data, i))
 -            break;
 +    struct mptcp_subflow_context *subflow;
 -        mptcp_subflow_set_scheduled(bpf_mptcp_subflow_ctx_by_pos(data, i), true);
 -    }
-+    bpf_for_each(mptcp_subflow, subflow, msk)
++    bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk)
 +        mptcp_subflow_set_scheduled(subflow, true);
      return 0;
  }
 --
 .43.0

-[PATCH mptcp-next v6 08/10] Squash to "selftests/bpf: Add bpf_burst scheduler & test"
+[PATCH mptcp-next v13 8/9] Squash to "selftests/bpf: Add bpf_burst scheduler & test"
 From: Geliang Tang <tanggeliang@kylinos.cn>
 Use the newly added bpf_for_each() helper to walk the conn_list.
+Drop bpf_subflow_send_info, use subflow_send_info instead.
 Drop mptcp_subflow_active declaration.
 Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
 ---
- .../selftests/bpf/progs/mptcp_bpf_burst.c     | 77 ++++++++++---------
+ .../selftests/bpf/progs/mptcp_bpf_burst.c     | 78 +++++++------------
-file changed, 39 insertions(+), 38 deletions(-)
+file changed, 26 insertions(+), 52 deletions(-)
 diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c
 index XXXXXXX..XXXXXXX 100644
 --- a/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c
 +++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c
 @@ -XXX,XX +XXX,XX @@ char _license[] SEC("license") = "GPL";
  #define min(a, b) ((a) < (b) ? (a) : (b))
-+#define SSK_MODE_ACTIVE    0
+-struct bpf_subflow_send_info {
-+#define SSK_MODE_BACKUP    1
+-    __u8 subflow_id;
-+#define SSK_MODE_MAX    2
+-    __u64 linger_time;
-+
+-};
- struct bpf_subflow_send_info {
+-
-     __u8 subflow_id;
+ extern bool mptcp_subflow_active(struct mptcp_subflow_context *subflow) __ksym;
-     __u64 linger_time;
+ extern void mptcp_set_timeout(struct sock *sk) __ksym;
-@@ -XXX,XX +XXX,XX @@ extern bool tcp_stream_memory_free(const struct sock *sk, int wake) __ksym;
+ extern __u64 mptcp_wnd_end(const struct mptcp_sock *msk) __ksym;
 -extern bool tcp_stream_memory_free(const struct sock *sk, int wake) __ksym;
 +extern bool bpf_sk_stream_memory_free(const struct sock *sk) __ksym;
  extern bool bpf_mptcp_subflow_queues_empty(struct sock *sk) __ksym;
  extern void mptcp_pm_subflow_chk_stale(const struct mptcp_sock *msk, struct sock *ssk) __ksym;
--#define SSK_MODE_ACTIVE    0
+@@ -XXX,XX +XXX,XX @@ static __always_inline bool tcp_rtx_and_write_queues_empty(struct sock *sk)
--#define SSK_MODE_BACKUP    1
+     return bpf_mptcp_subflow_queues_empty(sk) && tcp_write_queue_empty(sk);
--#define SSK_MODE_MAX    2
+ }
 -static __always_inline bool __sk_stream_memory_free(const struct sock *sk, int wake)
 -{
 -    if (sk->sk_wmem_queued >= sk->sk_sndbuf)
 -        return false;
 -
- static __always_inline __u64 div_u64(__u64 dividend, __u32 divisor)
+-    return tcp_stream_memory_free(sk, wake);
- {
+-}
-     return dividend / divisor;
+-
-@@ -XXX,XX +XXX,XX @@ static __always_inline bool sk_stream_memory_free(const struct sock *sk)
+-static __always_inline bool sk_stream_memory_free(const struct sock *sk)
-     return __sk_stream_memory_free(sk, 0);
+-{
- }
+-    return __sk_stream_memory_free(sk, 0);
+-}
-+static struct mptcp_subflow_context *
+-
 +mptcp_lookup_subflow_by_id(struct mptcp_sock *msk, unsigned int id)
 +{
 +    struct mptcp_subflow_context *subflow;
 +
 +    bpf_for_each(mptcp_subflow, subflow, msk) {
 +        if (subflow->subflow_id == id)
 +            return subflow;
 +    }
 +
 +    return NULL;
 +}
 +
  SEC("struct_ops")
  void BPF_PROG(mptcp_sched_burst_init, struct mptcp_sock *msk)
  {
-@@ -XXX,XX +XXX,XX @@ void BPF_PROG(mptcp_sched_burst_release, struct mptcp_sock *msk)
+@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
  int BPF_PROG(bpf_burst_get_send, struct mptcp_sock *msk,
           struct mptcp_sched_data *data)
  {
- }
+-    struct bpf_subflow_send_info send_info[SSK_MODE_MAX];
++    struct subflow_send_info send_info[SSK_MODE_MAX];
 -static int bpf_burst_get_send(struct mptcp_sock *msk,
 -                  struct mptcp_sched_data *data)
 +static int bpf_burst_get_send(struct mptcp_sock *msk)
  {
      struct bpf_subflow_send_info send_info[SSK_MODE_MAX];
      struct mptcp_subflow_context *subflow;
-@@ -XXX,XX +XXX,XX @@ static int bpf_burst_get_send(struct mptcp_sock *msk,
+     struct sock *sk = (struct sock *)msk;
      __u32 pace, burst, wmem;
@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_burst_get_send, struct mptcp_sock *msk,
      /* pick the subflow with the lower wmem/wspace ratio */
      for (i = 0; i < SSK_MODE_MAX; ++i) {
 -        send_info[i].subflow_id = MPTCP_SUBFLOWS_MAX;
 +        send_info[i].ssk = NULL;
          send_info[i].linger_time = -1;
      }
 -    for (i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
 -        bool backup;
-+    bpf_for_each(mptcp_subflow, subflow, msk) {
+-
 +        bool backup = subflow->backup || subflow->request_bkup;
 -        subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
 -        if (!subflow)
 -            break;
 -
 -        backup = subflow->backup || subflow->request_bkup;
--
++    bpf_for_each(mptcp_subflow, subflow, sk) {
--        ssk = mptcp_subflow_tcp_sock(subflow);
++        bool backup = subflow->backup || subflow->request_bkup;
-+        ssk = bpf_mptcp_subflow_tcp_sock(subflow);
          ssk = mptcp_subflow_tcp_sock(subflow);
          if (!mptcp_subflow_active(subflow))
-             continue;
+@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_burst_get_send, struct mptcp_sock *msk,
@@ -XXX,XX +XXX,XX @@ static int bpf_burst_get_send(struct mptcp_sock *msk,
          linger_time = div_u64((__u64)ssk->sk_wmem_queued << 32, pace);
          if (linger_time < send_info[backup].linger_time) {
 -            send_info[backup].subflow_id = i;
-+            send_info[backup].subflow_id = subflow->subflow_id;
++            send_info[backup].ssk = ssk;
              send_info[backup].linger_time = linger_time;
          }
      }
-@@ -XXX,XX +XXX,XX @@ static int bpf_burst_get_send(struct mptcp_sock *msk,
+@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_burst_get_send, struct mptcp_sock *msk,
      /* pick the best backup if no other subflow is active */
      if (!nr_active)
-         send_info[SSK_MODE_ACTIVE].subflow_id = send_info[SSK_MODE_BACKUP].subflow_id;
+-        send_info[SSK_MODE_ACTIVE].subflow_id = send_info[SSK_MODE_BACKUP].subflow_id;
 +        send_info[SSK_MODE_ACTIVE].ssk = send_info[SSK_MODE_BACKUP].ssk;
 -    subflow = bpf_mptcp_subflow_ctx_by_pos(data, send_info[SSK_MODE_ACTIVE].subflow_id);
-+    subflow = mptcp_lookup_subflow_by_id(msk, send_info[SSK_MODE_ACTIVE].subflow_id);
+-    if (!subflow)
-     if (!subflow)
++    ssk = send_info[SSK_MODE_ACTIVE].ssk;
 +    if (!ssk || !bpf_sk_stream_memory_free(ssk))
          return -1;
 -    ssk = mptcp_subflow_tcp_sock(subflow);
-+    ssk = bpf_mptcp_subflow_tcp_sock(subflow);
+-    if (!ssk || !sk_stream_memory_free(ssk))
-     if (!ssk || !sk_stream_memory_free(ssk))
++
 +    subflow = bpf_mptcp_subflow_ctx(ssk);
 +    if (!subflow)
          return -1;
-@@ -XXX,XX +XXX,XX @@ static int bpf_burst_get_send(struct mptcp_sock *msk,
+     burst = min(MPTCP_SEND_BURST_SIZE, mptcp_wnd_end(msk) - msk->snd_nxt);
-     return 0;
++    ssk = bpf_core_cast(ssk, struct sock);
- }
+     wmem = ssk->sk_wmem_queued;
+     if (!burst)
--static int bpf_burst_get_retrans(struct mptcp_sock *msk,
+         goto out;
--                 struct mptcp_sched_data *data)
+@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
-+static int bpf_burst_get_retrans(struct mptcp_sock *msk)
+ int BPF_PROG(bpf_burst_get_retrans, struct mptcp_sock *msk,
           struct mptcp_sched_data *data)
  {
 -    int backup = MPTCP_SUBFLOWS_MAX, pick = MPTCP_SUBFLOWS_MAX, subflow_id;
 +    struct sock *backup = NULL, *pick = NULL;
      struct mptcp_subflow_context *subflow;
      int min_stale_count = INT_MAX;
 -    struct sock *ssk;
 -    for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
 -        subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
 -        if (!subflow)
 -            break;
-+    bpf_for_each(mptcp_subflow, subflow, msk) {
++    bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk) {
 +        struct sock *ssk = bpf_mptcp_subflow_tcp_sock(subflow);
-         if (!mptcp_subflow_active(subflow))
+-        if (!mptcp_subflow_active(subflow))
 +        if (!ssk || !mptcp_subflow_active(subflow))
              continue;
 -        ssk = mptcp_subflow_tcp_sock(subflow);
          /* still data outstanding at TCP level? skip this */
          if (!tcp_rtx_and_write_queues_empty(ssk)) {
              mptcp_pm_subflow_chk_stale(msk, ssk);
-@@ -XXX,XX +XXX,XX @@ static int bpf_burst_get_retrans(struct mptcp_sock *msk,
+@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_burst_get_retrans, struct mptcp_sock *msk,
          }
          if (subflow->backup || subflow->request_bkup) {
 -            if (backup == MPTCP_SUBFLOWS_MAX)
 -                backup = i;
 ...
 +        return -1;
 +    subflow = bpf_mptcp_subflow_ctx(pick);
      if (!subflow)
          return -1;
      mptcp_subflow_set_scheduled(subflow, true);
-@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_burst_get_subflow, struct mptcp_sock *msk,
-          struct mptcp_sched_data *data)
- {
-     if (data->reinject)
--        return bpf_burst_get_retrans(msk, data);
--    return bpf_burst_get_send(msk, data);
-+        return bpf_burst_get_retrans(msk);
-+    return bpf_burst_get_send(msk);
- }
- SEC(".struct_ops")
 --
 .43.0

-[PATCH mptcp-next v6 09/10] mptcp: drop subflow contexts in mptcp_sched_data
+[PATCH mptcp-next v13 9/9] mptcp: drop subflow contexts in mptcp_sched_data
 ...
 The mptcp_subflow bpf_iter is added now, it's better to use the helper
 bpf_for_each(mptcp_subflow) to traverse all subflows on the conn_list of
 an MPTCP socket and then call kfunc to modify the fields of each subflow
 in the WIP MPTCP BPF packet scheduler examples, instead of converting them
 to a fixed array. With this helper, we can get rid of this subflow array
-"contexts" and the size of it "subflows" in struct mptcp_sched_data.
+"contexts" in struct mptcp_sched_data.
 Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
 ---
- include/net/mptcp.h | 2 --
+ include/net/mptcp.h | 3 ---
-file changed, 2 deletions(-)
+file changed, 3 deletions(-)
 diff --git a/include/net/mptcp.h b/include/net/mptcp.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/net/mptcp.h
 +++ b/include/net/mptcp.h
 @@ -XXX,XX +XXX,XX @@ struct mptcp_out_options {
+ #define MPTCP_SCHED_MAX        128
  #define MPTCP_SCHED_BUF_MAX    (MPTCP_SCHED_NAME_MAX * MPTCP_SCHED_MAX)
 -#define MPTCP_SUBFLOWS_MAX    8
 -
  struct mptcp_sched_data {
-     bool    reinject;
+     u8    subflows;
 -    u8    subflows;
 -    struct mptcp_subflow_context *contexts[MPTCP_SUBFLOWS_MAX];
  };
  struct mptcp_sched_ops {
 --
 .43.0

-[PATCH mptcp-next v6 10/10] Squash to "selftests/bpf: Add bpf scheduler test" - drop has_bytes_sent
+Deleted patch
-From: Geliang Tang <tanggeliang@kylinos.cn>
-Drop ss_search() and has_bytes_sent(), add a new bpf program to check
-the bytes_sent.
-Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
----
- .../testing/selftests/bpf/prog_tests/mptcp.c  | 48 ++++++++++---------
- .../selftests/bpf/progs/mptcp_bpf_bytes.c     | 39 +++++++++++++++
-files changed, 65 insertions(+), 22 deletions(-)
- create mode 100644 tools/testing/selftests/bpf/progs/mptcp_bpf_bytes.c
-diff --git a/tools/testing/selftests/bpf/prog_tests/mptcp.c b/tools/testing/selftests/bpf/prog_tests/mptcp.c
-index XXXXXXX..XXXXXXX 100644
---- a/tools/testing/selftests/bpf/prog_tests/mptcp.c
-+++ b/tools/testing/selftests/bpf/prog_tests/mptcp.c
-@@ -XXX,XX +XXX,XX @@
- #include "mptcpify.skel.h"
- #include "mptcp_subflow.skel.h"
- #include "mptcp_bpf_iters.skel.h"
-+#include "mptcp_bpf_bytes.skel.h"
- #include "mptcp_bpf_first.skel.h"
- #include "mptcp_bpf_bkup.skel.h"
- #include "mptcp_bpf_rr.skel.h"
-@@ -XXX,XX +XXX,XX @@ static struct nstoken *sched_init(char *flags, char *sched)
-     return NULL;
- }
--static int ss_search(char *src, char *dst, char *port, char *keyword)
--{
--    return SYS_NOFAIL("ip netns exec %s ss -enita src %s dst %s %s %d | grep -q '%s'",
--              NS_TEST, src, dst, port, PORT_1, keyword);
--}
--
--static int has_bytes_sent(char *dst)
--{
--    return ss_search(ADDR_1, dst, "sport", "bytes_sent:");
--}
--
- static void send_data_and_verify(char *sched, bool addr1, bool addr2)
- {
-+    int server_fd, client_fd, err;
-+    struct mptcp_bpf_bytes *skel;
-     struct timespec start, end;
--    int server_fd, client_fd;
-     unsigned int delta_ms;
-+    skel = mptcp_bpf_bytes__open_and_load();
-+    if (!ASSERT_OK_PTR(skel, "open_and_load: bytes"))
-+        return;
-+
-+    skel->bss->pid = getpid();
-+
-+    err = mptcp_bpf_bytes__attach(skel);
-+    if (!ASSERT_OK(err, "skel_attach: bytes"))
-+        goto skel_destroy;
-+
-     server_fd = start_mptcp_server(AF_INET, ADDR_1, PORT_1, 0);
-     if (!ASSERT_OK_FD(server_fd, "start_mptcp_server"))
--        return;
-+        goto skel_destroy;
-     client_fd = connect_to_fd(server_fd, 0);
-     if (!ASSERT_OK_FD(client_fd, "connect_to_fd"))
--        goto fail;
-+        goto close_server;
-     if (clock_gettime(CLOCK_MONOTONIC, &start) < 0)
--        goto fail;
-+        goto close_client;
-     if (!ASSERT_OK(send_recv_data(server_fd, client_fd, total_bytes),
-                "send_recv_data"))
--        goto fail;
-+        goto close_client;
-     if (clock_gettime(CLOCK_MONOTONIC, &end) < 0)
--        goto fail;
-+        goto close_client;
-     delta_ms = (end.tv_sec - start.tv_sec) * 1000 + (end.tv_nsec - start.tv_nsec) / 1000000;
-     printf("%s: %u ms\n", sched, delta_ms);
-     if (addr1)
--        CHECK(has_bytes_sent(ADDR_1), sched, "should have bytes_sent on addr1\n");
-+        ASSERT_GT(skel->bss->bytes_sent_1, 0, "should have bytes_sent on addr1");
-     else
--        CHECK(!has_bytes_sent(ADDR_1), sched, "shouldn't have bytes_sent on addr1\n");
-+        ASSERT_EQ(skel->bss->bytes_sent_1, 0, "shouldn't have bytes_sent on addr1");
-     if (addr2)
--        CHECK(has_bytes_sent(ADDR_2), sched, "should have bytes_sent on addr2\n");
-+        ASSERT_GT(skel->bss->bytes_sent_2, 0, "should have bytes_sent on addr2");
-     else
--        CHECK(!has_bytes_sent(ADDR_2), sched, "shouldn't have bytes_sent on addr2\n");
-+        ASSERT_EQ(skel->bss->bytes_sent_2, 0, "shouldn't have bytes_sent on addr2");
-+close_client:
-     close(client_fd);
--fail:
-+close_server:
-     close(server_fd);
-+skel_destroy:
-+    mptcp_bpf_bytes__destroy(skel);
- }
- static void test_default(void)
-diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_bytes.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_bytes.c
-new file mode 100644
-index XXXXXXX..XXXXXXX
---- /dev/null
-+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_bytes.c
-@@ -XXX,XX +XXX,XX @@
-+// SPDX-License-Identifier: GPL-2.0
-+/* Copyright (c) 2024, Kylin Software */
-+
-+/* vmlinux.h, bpf_helpers.h and other 'define' */
-+#include "bpf_tracing_net.h"
-+#include "mptcp_bpf.h"
-+
-+char _license[] SEC("license") = "GPL";
-+u64 bytes_sent_1 = 0;
-+u64 bytes_sent_2 = 0;
-+int pid;
-+
-+SEC("fexit/mptcp_sched_get_send")
-+int BPF_PROG(trace_mptcp_sched_get_send, struct mptcp_sock *msk)
-+{
-+    struct mptcp_subflow_context *subflow;
-+
-+    if (bpf_get_current_pid_tgid() >> 32 != pid)
-+        return 0;
-+
-+    if (!msk->pm.server_side)
-+        return 0;
-+
-+    mptcp_for_each_subflow(msk, subflow) {
-+        struct tcp_sock *tp;
-+        struct sock *ssk;
-+
-+        subflow = bpf_core_cast(subflow, struct mptcp_subflow_context);
-+        ssk = mptcp_subflow_tcp_sock(subflow);
-+        tp = bpf_core_cast(ssk, struct tcp_sock);
-+
-+        if (subflow->subflow_id == 1)
-+            bytes_sent_1 = tp->bytes_sent;
-+        else if (subflow->subflow_id == 2)
-+            bytes_sent_2 = tp->bytes_sent;
-+    }
-+
-+    return 0;
-+}
---
-.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

v6:
 - rebased to "add mptcp_subflow bpf_iter" v10

v5:
 - patch 2, drop mptcp_sock_type and mptcp_subflow_type.
 - patch 3, revert "bpf: Export more bpf_burst related functions"
 - patch 4, merge "bpf: Export more bpf_burst related functions" into it.

v4:
 - patch 2, a new cleanup for "bpf: Add bpf_mptcp_sched_ops".
 - patch 3 should be reverted.
 - patch 8, register kfunc_set.

v3:
 - rebased.
 - put the "drop has_bytes_sent" squash-to patch into this set.

v2:
 - update bpf_rr and bpf_burst

With the newly added mptcp_subflow bpf_iter, we can get rid of the
subflows array "contexts" in struct mptcp_sched_data. This set
uses bpf_for_each(mptcp_subflow) helper to update all the bpf
schedules:

bpf_for_each(mptcp_subflow, subflow, msk) {
                ... ...
                mptcp_subflow_set_scheduled(subflow, true);
        }

Depends on:
 - "add mptcp_subflow bpf_iter" v10

Based-on: <cover.1729063444.git.tanggeliang@kylinos.cn>

Geliang Tang (10):
  Revert "mptcp: add sched_data helpers"
  Squash to "bpf: Add bpf_mptcp_sched_ops"
  Squash to "bpf: Add bpf_mptcp_sched_kfunc_set"
  Squash to "selftests/bpf: Add bpf_first scheduler & test"
  Squash to "selftests/bpf: Add bpf_bkup scheduler & test"
  Squash to "selftests/bpf: Add bpf_rr scheduler & test"
  Squash to "selftests/bpf: Add bpf_red scheduler & test"
  Squash to "selftests/bpf: Add bpf_burst scheduler & test"
  mptcp: drop subflow contexts in mptcp_sched_data
  Squash to "selftests/bpf: Add bpf scheduler test" - drop
    has_bytes_sent

include/net/mptcp.h                           |  2 -
 net/mptcp/bpf.c                               | 24 ++----
 net/mptcp/protocol.h                          |  2 -
 net/mptcp/sched.c                             | 22 ------
 .../testing/selftests/bpf/prog_tests/mptcp.c  | 48 ++++++------
 tools/testing/selftests/bpf/progs/mptcp_bpf.h |  3 -
 .../selftests/bpf/progs/mptcp_bpf_bkup.c      | 16 +---
 .../selftests/bpf/progs/mptcp_bpf_burst.c     | 77 ++++++++++---------
 .../selftests/bpf/progs/mptcp_bpf_bytes.c     | 39 ++++++++++
 .../selftests/bpf/progs/mptcp_bpf_first.c     |  8 +-
 .../selftests/bpf/progs/mptcp_bpf_red.c       |  8 +-
 .../selftests/bpf/progs/mptcp_bpf_rr.c        | 24 +++---
 12 files changed, 133 insertions(+), 140 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/mptcp_bpf_bytes.c

-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Drop this patch.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/bpf.c      |  8 --------
 net/mptcp/protocol.h |  2 --
 net/mptcp/sched.c    | 22 ----------------------
 3 files changed, 32 deletions(-)

diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
index XXXXXXX..XXXXXXX 100644
--- a/net/mptcp/bpf.c
+++ b/net/mptcp/bpf.c
@@ -XXX,XX +XXX,XX @@ __bpf_kfunc static void bpf_mptcp_sock_release(struct mptcp_sock *msk)
 	WARN_ON_ONCE(!sk || !refcount_dec_not_one(&sk->sk_refcnt));
 }
 
-__bpf_kfunc struct mptcp_subflow_context *
-bpf_mptcp_subflow_ctx_by_pos(const struct mptcp_sched_data *data, unsigned int pos)
-{
-	if (pos >= MPTCP_SUBFLOWS_MAX)
-		return NULL;
-	return data->contexts[pos];
-}
-
 __bpf_kfunc static bool bpf_mptcp_subflow_queues_empty(struct sock *sk)
 {
 	return tcp_rtx_queue_empty(sk);
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index XXXXXXX..XXXXXXX 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -XXX,XX +XXX,XX @@ 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 mptcp_subflow_context *
-bpf_mptcp_subflow_ctx_by_pos(const struct mptcp_sched_data *data, unsigned int pos);
 struct sock *__mptcp_nmpc_sk(struct mptcp_sock *msk);
 bool __mptcp_close(struct sock *sk, long timeout);
 void mptcp_cancel_work(struct sock *sk);
diff --git a/net/mptcp/sched.c b/net/mptcp/sched.c
index XXXXXXX..XXXXXXX 100644
--- a/net/mptcp/sched.c
+++ b/net/mptcp/sched.c
@@ -XXX,XX +XXX,XX @@ void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 	WRITE_ONCE(subflow->scheduled, scheduled);
 }
 
-static void mptcp_sched_data_set_contexts(const struct mptcp_sock *msk,
-					  struct mptcp_sched_data *data)
-{
-	struct mptcp_subflow_context *subflow;
-	int i = 0;
-
-	mptcp_for_each_subflow(msk, subflow) {
-		if (i == MPTCP_SUBFLOWS_MAX) {
-			pr_warn_once("too many subflows");
-			break;
-		}
-		mptcp_subflow_set_scheduled(subflow, false);
-		data->contexts[i++] = subflow;
-	}
-	data->subflows = i;
-
-	for (; i < MPTCP_SUBFLOWS_MAX; i++)
-		data->contexts[i] = NULL;
-}
-
 int mptcp_sched_get_send(struct mptcp_sock *msk)
 {
 	struct mptcp_subflow_context *subflow;
@@ -XXX,XX +XXX,XX @@ int mptcp_sched_get_send(struct mptcp_sock *msk)
 	data.reinject = false;
 	if (msk->sched == &mptcp_sched_default || !msk->sched)
 		return mptcp_sched_default_get_subflow(msk, &data);
-	mptcp_sched_data_set_contexts(msk, &data);
 	return msk->sched->get_subflow(msk, &data);
 }
 
@@ -XXX,XX +XXX,XX @@ int mptcp_sched_get_retrans(struct mptcp_sock *msk)
 	data.reinject = true;
 	if (msk->sched == &mptcp_sched_default || !msk->sched)
 		return mptcp_sched_default_get_subflow(msk, &data);
-	mptcp_sched_data_set_contexts(msk, &data);
 	return msk->sched->get_subflow(msk, &data);
 }
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Please update the subject to

bpf: Add mptcp packet scheduler struct_ops

Drop mptcp_sock_type and mptcp_subflow_type.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/bpf.c | 15 ++++++---------
 1 file changed, 6 insertions(+), 9 deletions(-)

diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
index XXXXXXX..XXXXXXX 100644
--- a/net/mptcp/bpf.c
+++ b/net/mptcp/bpf.c
@@ -XXX,XX +XXX,XX @@
 
 #ifdef CONFIG_BPF_JIT
 static struct bpf_struct_ops bpf_mptcp_sched_ops;
-static const struct btf_type *mptcp_sock_type, *mptcp_subflow_type __read_mostly;
 static u32 mptcp_sock_id, mptcp_subflow_id;
 
+/* MPTCP BPF packet scheduler */
+
 static const struct bpf_func_proto *
 bpf_mptcp_sched_get_func_proto(enum bpf_func_id func_id,
 			       const struct bpf_prog *prog)
@@ -XXX,XX +XXX,XX @@ static int bpf_mptcp_sched_btf_struct_access(struct bpf_verifier_log *log,
 					     const struct bpf_reg_state *reg,
 					     int off, int size)
 {
-	const struct btf_type *t;
+	u32 id = reg->btf_id;
 	size_t end;
 
-	t = btf_type_by_id(reg->btf, reg->btf_id);
-
-	if (t == mptcp_sock_type) {
+	if (id == mptcp_sock_id) {
 		switch (off) {
 		case offsetof(struct mptcp_sock, snd_burst):
 			end = offsetofend(struct mptcp_sock, snd_burst);
@@ -XXX,XX +XXX,XX @@ static int bpf_mptcp_sched_btf_struct_access(struct bpf_verifier_log *log,
 				off);
 			return -EACCES;
 		}
-	} else if (t == mptcp_subflow_type) {
+	} else if (id == mptcp_subflow_id) {
 		switch (off) {
 		case offsetof(struct mptcp_subflow_context, avg_pacing_rate):
 			end = offsetofend(struct mptcp_subflow_context, avg_pacing_rate);
@@ -XXX,XX +XXX,XX @@ static int bpf_mptcp_sched_btf_struct_access(struct bpf_verifier_log *log,
 
 	if (off + size > end) {
 		bpf_log(log, "access beyond %s at off %u size %u ended at %zu",
-			t == mptcp_sock_type ? "mptcp_sock" : "mptcp_subflow_context",
+			id == mptcp_sock_id ? "mptcp_sock" : "mptcp_subflow_context",
 			off, size, end);
 		return -EACCES;
 	}
@@ -XXX,XX +XXX,XX @@ static int bpf_mptcp_sched_init(struct btf *btf)
 	if (type_id < 0)
 		return -EINVAL;
 	mptcp_sock_id = type_id;
-	mptcp_sock_type = btf_type_by_id(btf, mptcp_sock_id);
 
 	type_id = btf_find_by_name_kind(btf, "mptcp_subflow_context",
 					BTF_KIND_STRUCT);
 	if (type_id < 0)
 		return -EINVAL;
 	mptcp_subflow_id = type_id;
-	mptcp_subflow_type = btf_type_by_id(btf, mptcp_subflow_id);
 
 	return 0;
 }
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Use the newly added bpf_for_each() helper to walk the conn_list.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/bpf/progs/mptcp_bpf.h       | 3 ---
 tools/testing/selftests/bpf/progs/mptcp_bpf_first.c | 8 +++++++-
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf.h b/tools/testing/selftests/bpf/progs/mptcp_bpf.h
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf.h
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf.h
@@ -XXX,XX +XXX,XX @@ bpf_mptcp_subflow_tcp_sock(const struct mptcp_subflow_context *subflow) __ksym;
 extern void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 					bool scheduled) __ksym;
 
-extern struct mptcp_subflow_context *
-bpf_mptcp_subflow_ctx_by_pos(const struct mptcp_sched_data *data, unsigned int pos) __ksym;
-
 #endif
diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
 int BPF_PROG(bpf_first_get_subflow, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
-	mptcp_subflow_set_scheduled(bpf_mptcp_subflow_ctx_by_pos(data, 0), true);
+	struct mptcp_subflow_context *subflow;
+
+	bpf_for_each(mptcp_subflow, subflow, msk) {
+		mptcp_subflow_set_scheduled(subflow, true);
+		break;
+	}
+
 	return 0;
 }
 
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Use the newly added bpf_for_each() helper to walk the conn_list.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../testing/selftests/bpf/progs/mptcp_bpf_bkup.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
 int BPF_PROG(bpf_bkup_get_subflow, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
-	int nr = -1;
-
-	for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
-		struct mptcp_subflow_context *subflow;
-
-		subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
-		if (!subflow)
-			break;
+	struct mptcp_subflow_context *subflow;
 
+	bpf_for_each(mptcp_subflow, subflow, msk) {
 		if (!BPF_CORE_READ_BITFIELD_PROBED(subflow, backup) ||
 		    !BPF_CORE_READ_BITFIELD_PROBED(subflow, request_bkup)) {
-			nr = i;
+			mptcp_subflow_set_scheduled(subflow, true);
 			break;
 		}
 	}
 
-	if (nr != -1) {
-		mptcp_subflow_set_scheduled(bpf_mptcp_subflow_ctx_by_pos(data, nr), true);
-		return -1;
-	}
 	return 0;
 }
 
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Use the newly added bpf_for_each() helper to walk the conn_list.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/bpf/progs/mptcp_bpf_rr.c        | 24 ++++++++-----------
 1 file changed, 10 insertions(+), 14 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
 int BPF_PROG(bpf_rr_get_subflow, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
-	struct mptcp_subflow_context *subflow;
+	struct mptcp_subflow_context *subflow, *next;
 	struct mptcp_rr_storage *ptr;
 	struct sock *last_snd = NULL;
-	int nr = 0;
 
 	ptr = bpf_sk_storage_get(&mptcp_rr_map, msk, 0,
 				 BPF_LOCAL_STORAGE_GET_F_CREATE);
@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_rr_get_subflow, struct mptcp_sock *msk,
 		return -1;
 
 	last_snd = ptr->last_snd;
+	next = bpf_mptcp_subflow_ctx(msk->first);
 
-	for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
-		subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
-		if (!last_snd || !subflow)
+	bpf_for_each(mptcp_subflow, subflow, msk) {
+		if (!last_snd)
 			break;
 
-		if (mptcp_subflow_tcp_sock(subflow) == last_snd) {
-			if (i + 1 == MPTCP_SUBFLOWS_MAX ||
-			    !bpf_mptcp_subflow_ctx_by_pos(data, i + 1))
+		if (bpf_mptcp_subflow_tcp_sock(subflow) == last_snd) {
+			subflow = bpf_iter_mptcp_subflow_next(&___it);
+			if (!subflow)
 				break;
 
-			nr = i + 1;
+			next = subflow;
 			break;
 		}
 	}
 
-	subflow = bpf_mptcp_subflow_ctx_by_pos(data, nr);
-	if (!subflow)
-		return -1;
-	mptcp_subflow_set_scheduled(subflow, true);
-	ptr->last_snd = mptcp_subflow_tcp_sock(subflow);
+	mptcp_subflow_set_scheduled(next, true);
+	ptr->last_snd = bpf_mptcp_subflow_tcp_sock(next);
 	return 0;
 }
 
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Use the newly added bpf_for_each() helper to walk the conn_list.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/bpf/progs/mptcp_bpf_red.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
 int BPF_PROG(bpf_red_get_subflow, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
-	for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
-		if (!bpf_mptcp_subflow_ctx_by_pos(data, i))
-			break;
+	struct mptcp_subflow_context *subflow;
 
-		mptcp_subflow_set_scheduled(bpf_mptcp_subflow_ctx_by_pos(data, i), true);
-	}
+	bpf_for_each(mptcp_subflow, subflow, msk)
+		mptcp_subflow_set_scheduled(subflow, true);
 
 	return 0;
 }
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Use the newly added bpf_for_each() helper to walk the conn_list.

Drop mptcp_subflow_active declaration.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/bpf/progs/mptcp_bpf_burst.c     | 77 ++++++++++---------
 1 file changed, 39 insertions(+), 38 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c
@@ -XXX,XX +XXX,XX @@ char _license[] SEC("license") = "GPL";
 
 #define min(a, b) ((a) < (b) ? (a) : (b))
 
+#define SSK_MODE_ACTIVE	0
+#define SSK_MODE_BACKUP	1
+#define SSK_MODE_MAX	2
+
 struct bpf_subflow_send_info {
 	__u8 subflow_id;
 	__u64 linger_time;
@@ -XXX,XX +XXX,XX @@ extern bool tcp_stream_memory_free(const struct sock *sk, int wake) __ksym;
 extern bool bpf_mptcp_subflow_queues_empty(struct sock *sk) __ksym;
 extern void mptcp_pm_subflow_chk_stale(const struct mptcp_sock *msk, struct sock *ssk) __ksym;
 
-#define SSK_MODE_ACTIVE	0
-#define SSK_MODE_BACKUP	1
-#define SSK_MODE_MAX	2
-
 static __always_inline __u64 div_u64(__u64 dividend, __u32 divisor)
 {
 	return dividend / divisor;
@@ -XXX,XX +XXX,XX @@ static __always_inline bool sk_stream_memory_free(const struct sock *sk)
 	return __sk_stream_memory_free(sk, 0);
 }
 
+static struct mptcp_subflow_context *
+mptcp_lookup_subflow_by_id(struct mptcp_sock *msk, unsigned int id)
+{
+	struct mptcp_subflow_context *subflow;
+
+	bpf_for_each(mptcp_subflow, subflow, msk) {
+		if (subflow->subflow_id == id)
+			return subflow;
+	}
+
+	return NULL;
+}
+
 SEC("struct_ops")
 void BPF_PROG(mptcp_sched_burst_init, struct mptcp_sock *msk)
 {
@@ -XXX,XX +XXX,XX @@ void BPF_PROG(mptcp_sched_burst_release, struct mptcp_sock *msk)
 {
 }
 
-static int bpf_burst_get_send(struct mptcp_sock *msk,
-			      struct mptcp_sched_data *data)
+static int bpf_burst_get_send(struct mptcp_sock *msk)
 {
 	struct bpf_subflow_send_info send_info[SSK_MODE_MAX];
 	struct mptcp_subflow_context *subflow;
@@ -XXX,XX +XXX,XX @@ static int bpf_burst_get_send(struct mptcp_sock *msk,
 		send_info[i].linger_time = -1;
 	}
 
-	for (i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
-		bool backup;
+	bpf_for_each(mptcp_subflow, subflow, msk) {
+		bool backup = subflow->backup || subflow->request_bkup;
 
-		subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
-		if (!subflow)
-			break;
-
-		backup = subflow->backup || subflow->request_bkup;
-
-		ssk = mptcp_subflow_tcp_sock(subflow);
+		ssk = bpf_mptcp_subflow_tcp_sock(subflow);
 		if (!mptcp_subflow_active(subflow))
 			continue;
 
@@ -XXX,XX +XXX,XX @@ static int bpf_burst_get_send(struct mptcp_sock *msk,
 
 		linger_time = div_u64((__u64)ssk->sk_wmem_queued << 32, pace);
 		if (linger_time < send_info[backup].linger_time) {
-			send_info[backup].subflow_id = i;
+			send_info[backup].subflow_id = subflow->subflow_id;
 			send_info[backup].linger_time = linger_time;
 		}
 	}
@@ -XXX,XX +XXX,XX @@ static int bpf_burst_get_send(struct mptcp_sock *msk,
 	if (!nr_active)
 		send_info[SSK_MODE_ACTIVE].subflow_id = send_info[SSK_MODE_BACKUP].subflow_id;
 
-	subflow = bpf_mptcp_subflow_ctx_by_pos(data, send_info[SSK_MODE_ACTIVE].subflow_id);
+	subflow = mptcp_lookup_subflow_by_id(msk, send_info[SSK_MODE_ACTIVE].subflow_id);
 	if (!subflow)
 		return -1;
-	ssk = mptcp_subflow_tcp_sock(subflow);
+	ssk = bpf_mptcp_subflow_tcp_sock(subflow);
 	if (!ssk || !sk_stream_memory_free(ssk))
 		return -1;
 
@@ -XXX,XX +XXX,XX @@ static int bpf_burst_get_send(struct mptcp_sock *msk,
 	return 0;
 }
 
-static int bpf_burst_get_retrans(struct mptcp_sock *msk,
-				 struct mptcp_sched_data *data)
+static int bpf_burst_get_retrans(struct mptcp_sock *msk)
 {
-	int backup = MPTCP_SUBFLOWS_MAX, pick = MPTCP_SUBFLOWS_MAX, subflow_id;
+	struct sock *backup = NULL, *pick = NULL;
 	struct mptcp_subflow_context *subflow;
 	int min_stale_count = INT_MAX;
-	struct sock *ssk;
 
-	for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
-		subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
-		if (!subflow)
-			break;
+	bpf_for_each(mptcp_subflow, subflow, msk) {
+		struct sock *ssk = bpf_mptcp_subflow_tcp_sock(subflow);
 
 		if (!mptcp_subflow_active(subflow))
 			continue;
 
-		ssk = mptcp_subflow_tcp_sock(subflow);
 		/* still data outstanding at TCP level? skip this */
 		if (!tcp_rtx_and_write_queues_empty(ssk)) {
 			mptcp_pm_subflow_chk_stale(msk, ssk);
@@ -XXX,XX +XXX,XX @@ static int bpf_burst_get_retrans(struct mptcp_sock *msk,
 		}
 
 		if (subflow->backup || subflow->request_bkup) {
-			if (backup == MPTCP_SUBFLOWS_MAX)
-				backup = i;
+			if (!backup)
+				backup = ssk;
 			continue;
 		}
 
-		if (pick == MPTCP_SUBFLOWS_MAX)
-			pick = i;
+		if (!pick)
+			pick = ssk;
 	}
 
-	if (pick < MPTCP_SUBFLOWS_MAX) {
-		subflow_id = pick;
+	if (pick)
 		goto out;
-	}
-	subflow_id = min_stale_count > 1 ? backup : MPTCP_SUBFLOWS_MAX;
+	pick = min_stale_count > 1 ? backup : NULL;
 
 out:
-	subflow = bpf_mptcp_subflow_ctx_by_pos(data, subflow_id);
+	if (!pick)
+		return -1;
+	subflow = bpf_mptcp_subflow_ctx(pick);
 	if (!subflow)
 		return -1;
 	mptcp_subflow_set_scheduled(subflow, true);
@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_burst_get_subflow, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
 	if (data->reinject)
-		return bpf_burst_get_retrans(msk, data);
-	return bpf_burst_get_send(msk, data);
+		return bpf_burst_get_retrans(msk);
+	return bpf_burst_get_send(msk);
 }
 
 SEC(".struct_ops")
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Drop ss_search() and has_bytes_sent(), add a new bpf program to check
the bytes_sent.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../testing/selftests/bpf/prog_tests/mptcp.c  | 48 ++++++++++---------
 .../selftests/bpf/progs/mptcp_bpf_bytes.c     | 39 +++++++++++++++
 2 files changed, 65 insertions(+), 22 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/mptcp_bpf_bytes.c

diff --git a/tools/testing/selftests/bpf/prog_tests/mptcp.c b/tools/testing/selftests/bpf/prog_tests/mptcp.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/prog_tests/mptcp.c
+++ b/tools/testing/selftests/bpf/prog_tests/mptcp.c
@@ -XXX,XX +XXX,XX @@
 #include "mptcpify.skel.h"
 #include "mptcp_subflow.skel.h"
 #include "mptcp_bpf_iters.skel.h"
+#include "mptcp_bpf_bytes.skel.h"
 #include "mptcp_bpf_first.skel.h"
 #include "mptcp_bpf_bkup.skel.h"
 #include "mptcp_bpf_rr.skel.h"
@@ -XXX,XX +XXX,XX @@ static struct nstoken *sched_init(char *flags, char *sched)
 	return NULL;
 }
 
-static int ss_search(char *src, char *dst, char *port, char *keyword)
-{
-	return SYS_NOFAIL("ip netns exec %s ss -enita src %s dst %s %s %d | grep -q '%s'",
-			  NS_TEST, src, dst, port, PORT_1, keyword);
-}
-
-static int has_bytes_sent(char *dst)
-{
-	return ss_search(ADDR_1, dst, "sport", "bytes_sent:");
-}
-
 static void send_data_and_verify(char *sched, bool addr1, bool addr2)
 {
+	int server_fd, client_fd, err;
+	struct mptcp_bpf_bytes *skel;
 	struct timespec start, end;
-	int server_fd, client_fd;
 	unsigned int delta_ms;
 
+	skel = mptcp_bpf_bytes__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "open_and_load: bytes"))
+		return;
+
+	skel->bss->pid = getpid();
+
+	err = mptcp_bpf_bytes__attach(skel);
+	if (!ASSERT_OK(err, "skel_attach: bytes"))
+		goto skel_destroy;
+
 	server_fd = start_mptcp_server(AF_INET, ADDR_1, PORT_1, 0);
 	if (!ASSERT_OK_FD(server_fd, "start_mptcp_server"))
-		return;
+		goto skel_destroy;
 
 	client_fd = connect_to_fd(server_fd, 0);
 	if (!ASSERT_OK_FD(client_fd, "connect_to_fd"))
-		goto fail;
+		goto close_server;
 
 	if (clock_gettime(CLOCK_MONOTONIC, &start) < 0)
-		goto fail;
+		goto close_client;
 
 	if (!ASSERT_OK(send_recv_data(server_fd, client_fd, total_bytes),
 		       "send_recv_data"))
-		goto fail;
+		goto close_client;
 
 	if (clock_gettime(CLOCK_MONOTONIC, &end) < 0)
-		goto fail;
+		goto close_client;
 
 	delta_ms = (end.tv_sec - start.tv_sec) * 1000 + (end.tv_nsec - start.tv_nsec) / 1000000;
 	printf("%s: %u ms\n", sched, delta_ms);
 
 	if (addr1)
-		CHECK(has_bytes_sent(ADDR_1), sched, "should have bytes_sent on addr1\n");
+		ASSERT_GT(skel->bss->bytes_sent_1, 0, "should have bytes_sent on addr1");
 	else
-		CHECK(!has_bytes_sent(ADDR_1), sched, "shouldn't have bytes_sent on addr1\n");
+		ASSERT_EQ(skel->bss->bytes_sent_1, 0, "shouldn't have bytes_sent on addr1");
 	if (addr2)
-		CHECK(has_bytes_sent(ADDR_2), sched, "should have bytes_sent on addr2\n");
+		ASSERT_GT(skel->bss->bytes_sent_2, 0, "should have bytes_sent on addr2");
 	else
-		CHECK(!has_bytes_sent(ADDR_2), sched, "shouldn't have bytes_sent on addr2\n");
+		ASSERT_EQ(skel->bss->bytes_sent_2, 0, "shouldn't have bytes_sent on addr2");
 
+close_client:
 	close(client_fd);
-fail:
+close_server:
 	close(server_fd);
+skel_destroy:
+	mptcp_bpf_bytes__destroy(skel);
 }
 
 static void test_default(void)
diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_bytes.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_bytes.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_bytes.c
@@ -XXX,XX +XXX,XX @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2024, Kylin Software */
+
+/* vmlinux.h, bpf_helpers.h and other 'define' */
+#include "bpf_tracing_net.h"
+#include "mptcp_bpf.h"
+
+char _license[] SEC("license") = "GPL";
+u64 bytes_sent_1 = 0;
+u64 bytes_sent_2 = 0;
+int pid;
+
+SEC("fexit/mptcp_sched_get_send")
+int BPF_PROG(trace_mptcp_sched_get_send, struct mptcp_sock *msk)
+{
+	struct mptcp_subflow_context *subflow;
+
+	if (bpf_get_current_pid_tgid() >> 32 != pid)
+		return 0;
+
+	if (!msk->pm.server_side)
+		return 0;
+
+	mptcp_for_each_subflow(msk, subflow) {
+		struct tcp_sock *tp;
+		struct sock *ssk;
+
+		subflow = bpf_core_cast(subflow, struct mptcp_subflow_context);
+		ssk = mptcp_subflow_tcp_sock(subflow);
+		tp = bpf_core_cast(ssk, struct tcp_sock);
+
+		if (subflow->subflow_id == 1)
+			bytes_sent_1 = tp->bytes_sent;
+		else if (subflow->subflow_id == 2)
+			bytes_sent_2 = tp->bytes_sent;
+	}
+
+	return 0;
+}
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

v13:
 - use '__ign' suffix to ignore the argument type checks of
   bpf_mptcp_subflow_ctx() and bpf_sk_stream_memory_free(),
   instead of adding a new helper bpf_mptcp_send_info_to_ssk().
 - use 'bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk)' instead
   of using 'bpf_for_each(mptcp_subflow, subflow, msk)'.
 - keep struct mptcp_sched_data for future use.

Depends on:
 - Squash to "Add mptcp_subflow bpf_iter support", v2

Based-on: <cover.1738470660.git.tanggeliang@kylinos.cn>

v12:
 - drop struct mptcp_sched_data.
 - rebased on "split get_subflow interface into two" v2.

v11:
If another squash-to patchset (Squash to "Add mptcp_subflow bpf_iter
support") under review is merged before this set, v10 will fail to run.
v11 fixes this issue and can run regardless of whether it is merged
before or after the squash-to patchset.

Compared with v10, only patches 3, 5, and 8 have been modified:
 - use mptcp_subflow_tcp_sock instead of bpf_mptcp_subflow_tcp_sock in
   patch 3 and patch 5.
 - drop bpf_mptcp_sched_kfunc_set, use bpf_mptcp_common_kfunc_set instead
   in patch 8.

v10:
 - drop mptcp_subflow_set_scheduled() helper and WRITE_ONCE() in BPF.
 - add new bpf helper bpf_mptcp_send_info_to_ssk() for burst scheduler.

v9:
 - merge 'Fixes for "use bpf_iter in bpf schedulers" v8' into this set.
 - rebased on "add netns helpers" v4

v8:
 - address Mat's comments in v7.
 - move sk_stream_memory_free check inside bpf_for_each() loop.
 - implement mptcp_subflow_set_scheduled helper in BPF.
 - add cleanup patches into this set again.

v7:
 - move cleanup patches out of this set.
 - rebased.

v6:
 - rebased to "add mptcp_subflow bpf_iter" v10

v4:
 - patch 2, a new cleanup for "bpf: Add bpf_mptcp_sched_ops".
 - patch 3 should be reverted.
 - patch 8, register kfunc_set.

v3:
 - rebased.
 - put the "drop has_bytes_sent" squash-to patch into this set.

v2:
 - update bpf_rr and bpf_burst

bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk) {
                ... ...
                mptcp_subflow_set_scheduled(subflow, true);
        }

Geliang Tang (9):
  Squash to "bpf: Register mptcp common kfunc set"
  Revert "mptcp: add sched_data helpers"
  Squash to "bpf: Export mptcp packet scheduler helpers"
  Squash to "selftests/bpf: Add bpf_first scheduler & test"
  Squash to "selftests/bpf: Add bpf_bkup scheduler & test"
  Squash to "selftests/bpf: Add bpf_rr scheduler & test"
  Squash to "selftests/bpf: Add bpf_red scheduler & test"
  Squash to "selftests/bpf: Add bpf_burst scheduler & test"
  mptcp: drop subflow contexts in mptcp_sched_data

include/net/mptcp.h                           |  3 -
 net/mptcp/bpf.c                               | 52 +++++++------
 net/mptcp/protocol.h                          |  2 -
 net/mptcp/sched.c                             | 22 ------
 tools/testing/selftests/bpf/progs/mptcp_bpf.h |  3 -
 .../selftests/bpf/progs/mptcp_bpf_bkup.c      | 16 +---
 .../selftests/bpf/progs/mptcp_bpf_burst.c     | 78 +++++++------------
 .../selftests/bpf/progs/mptcp_bpf_first.c     |  8 +-
 .../selftests/bpf/progs/mptcp_bpf_red.c       |  8 +-
 .../selftests/bpf/progs/mptcp_bpf_rr.c        | 31 ++++----
 10 files changed, 80 insertions(+), 143 deletions(-)

-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Instead of adding a new BPF function bpf_mptcp_send_info_to_ssk() in
v12, this patch uses a much more simpler approach, which using '__ign'
suffix for the argument of bpf_mptcp_subflow_ctx() to let BPF to
ignore the type check of this argument.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/bpf.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
index XXXXXXX..XXXXXXX 100644
--- a/net/mptcp/bpf.c
+++ b/net/mptcp/bpf.c
@@ -XXX,XX +XXX,XX @@ struct bpf_iter_mptcp_subflow_kern {
 __bpf_kfunc_start_defs();
 
 __bpf_kfunc static struct mptcp_subflow_context *
-bpf_mptcp_subflow_ctx(const struct sock *sk)
+bpf_mptcp_subflow_ctx(const struct sock *sk__ign)
 {
-	if (sk && sk_fullsock(sk) &&
-	    sk->sk_protocol == IPPROTO_TCP && sk_is_mptcp(sk))
-		return mptcp_subflow_ctx(sk);
+	if (sk__ign && sk_fullsock(sk__ign) &&
+	    sk__ign->sk_protocol == IPPROTO_TCP && sk_is_mptcp(sk__ign))
+		return mptcp_subflow_ctx(sk__ign);
 
 	return NULL;
 }
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Drop this patch. bpf_mptcp_subflow_ctx_by_pos and
mptcp_sched_data_set_contexts are uesless now.

diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
index XXXXXXX..XXXXXXX 100644
--- a/net/mptcp/bpf.c
+++ b/net/mptcp/bpf.c
@@ -XXX,XX +XXX,XX @@ 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)
-{
-	if (pos >= MPTCP_SUBFLOWS_MAX)
-		return NULL;
-	return data->contexts[pos];
-}
-
 __bpf_kfunc static bool bpf_mptcp_subflow_queues_empty(struct sock *sk)
 {
 	return tcp_rtx_queue_empty(sk);
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index XXXXXXX..XXXXXXX 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -XXX,XX +XXX,XX @@ 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 mptcp_subflow_context *
-bpf_mptcp_subflow_ctx_by_pos(const struct mptcp_sched_data *data, unsigned int pos);
 struct sock *__mptcp_nmpc_sk(struct mptcp_sock *msk);
 bool __mptcp_close(struct sock *sk, long timeout);
 void mptcp_cancel_work(struct sock *sk);
diff --git a/net/mptcp/sched.c b/net/mptcp/sched.c
index XXXXXXX..XXXXXXX 100644
--- a/net/mptcp/sched.c
+++ b/net/mptcp/sched.c
@@ -XXX,XX +XXX,XX @@ void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 	WRITE_ONCE(subflow->scheduled, scheduled);
 }
 
-static void mptcp_sched_data_set_contexts(const struct mptcp_sock *msk,
-					  struct mptcp_sched_data *data)
-{
-	struct mptcp_subflow_context *subflow;
-	int i = 0;
-
-	mptcp_for_each_subflow(msk, subflow) {
-		if (i == MPTCP_SUBFLOWS_MAX) {
-			pr_warn_once("too many subflows");
-			break;
-		}
-		mptcp_subflow_set_scheduled(subflow, false);
-		data->contexts[i++] = subflow;
-	}
-	data->subflows = i;
-
-	for (; i < MPTCP_SUBFLOWS_MAX; i++)
-		data->contexts[i] = NULL;
-}
-
 int mptcp_sched_get_send(struct mptcp_sock *msk)
 {
 	struct mptcp_subflow_context *subflow;
@@ -XXX,XX +XXX,XX @@ int mptcp_sched_get_send(struct mptcp_sock *msk)
 
 	if (msk->sched == &mptcp_sched_default || !msk->sched)
 		return mptcp_sched_default_get_send(msk, &data);
-	mptcp_sched_data_set_contexts(msk, &data);
 	return msk->sched->get_send(msk, &data);
 }
 
@@ -XXX,XX +XXX,XX @@ int mptcp_sched_get_retrans(struct mptcp_sock *msk)
 	if (msk->sched == &mptcp_sched_default || !msk->sched)
 		return mptcp_sched_default_get_retrans(msk, &data);
 
-	mptcp_sched_data_set_contexts(msk, &data);
 	if (msk->sched->get_retrans)
 		return msk->sched->get_retrans(msk, &data);
 	return msk->sched->get_send(msk, &data);
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Remove bpf_mptcp_subflow_ctx_by_pos from BPF kfunc set.
Drop bpf_mptcp_sched_kfunc_set, use bpf_mptcp_common_kfunc_set instead.
Add new helpers bpf_mptcp_subflow_tcp_sock() and
bpf_sk_stream_memory_free().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/bpf.c | 38 ++++++++++++++++++++++++--------------
 1 file changed, 24 insertions(+), 14 deletions(-)

diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c
index XXXXXXX..XXXXXXX 100644
--- a/net/mptcp/bpf.c
+++ b/net/mptcp/bpf.c
@@ -XXX,XX +XXX,XX @@ bpf_mptcp_subflow_ctx(const struct sock *sk__ign)
 	return NULL;
 }
 
+__bpf_kfunc static struct sock *
+bpf_mptcp_subflow_tcp_sock(const struct mptcp_subflow_context *subflow)
+{
+	if (!subflow)
+		return NULL;
+
+	return mptcp_subflow_tcp_sock(subflow);
+}
+
 __bpf_kfunc static int
 bpf_iter_mptcp_subflow_new(struct bpf_iter_mptcp_subflow *it,
 			   struct sock *sk)
@@ -XXX,XX +XXX,XX @@ __bpf_kfunc static bool bpf_mptcp_subflow_queues_empty(struct sock *sk)
 	return tcp_rtx_queue_empty(sk);
 }
 
+__bpf_kfunc static bool bpf_sk_stream_memory_free(const struct sock *sk__ign)
+{
+	if (sk__ign && sk_fullsock(sk__ign) &&
+	    sk__ign->sk_protocol == IPPROTO_TCP && sk_is_mptcp(sk__ign))
+		return sk_stream_memory_free(sk__ign);
+
+	return NULL;
+}
+
 __bpf_kfunc_end_defs();
 
 BTF_KFUNCS_START(bpf_mptcp_common_kfunc_ids)
 BTF_ID_FLAGS(func, bpf_mptcp_subflow_ctx, KF_RET_NULL)
+BTF_ID_FLAGS(func, bpf_mptcp_subflow_tcp_sock, KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_iter_mptcp_subflow_new, KF_ITER_NEW | KF_TRUSTED_ARGS)
 BTF_ID_FLAGS(func, bpf_iter_mptcp_subflow_next, KF_ITER_NEXT | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_iter_mptcp_subflow_destroy, KF_ITER_DESTROY)
-BTF_KFUNCS_END(bpf_mptcp_common_kfunc_ids)
-
-static const struct btf_kfunc_id_set bpf_mptcp_common_kfunc_set = {
-	.owner	= THIS_MODULE,
-	.set	= &bpf_mptcp_common_kfunc_ids,
-};
-
-BTF_KFUNCS_START(bpf_mptcp_sched_kfunc_ids)
 BTF_ID_FLAGS(func, mptcp_subflow_set_scheduled)
-BTF_ID_FLAGS(func, bpf_mptcp_subflow_ctx_by_pos)
 BTF_ID_FLAGS(func, mptcp_subflow_active)
 BTF_ID_FLAGS(func, mptcp_set_timeout)
 BTF_ID_FLAGS(func, mptcp_wnd_end)
-BTF_ID_FLAGS(func, tcp_stream_memory_free)
+BTF_ID_FLAGS(func, bpf_sk_stream_memory_free, KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_mptcp_subflow_queues_empty)
 BTF_ID_FLAGS(func, mptcp_pm_subflow_chk_stale, KF_SLEEPABLE)
-BTF_KFUNCS_END(bpf_mptcp_sched_kfunc_ids)
+BTF_KFUNCS_END(bpf_mptcp_common_kfunc_ids)
 
-static const struct btf_kfunc_id_set bpf_mptcp_sched_kfunc_set = {
+static const struct btf_kfunc_id_set bpf_mptcp_common_kfunc_set = {
 	.owner	= THIS_MODULE,
-	.set	= &bpf_mptcp_sched_kfunc_ids,
+	.set	= &bpf_mptcp_common_kfunc_ids,
 };
 
 static int __init bpf_mptcp_kfunc_init(void)
@@ -XXX,XX +XXX,XX @@ static int __init bpf_mptcp_kfunc_init(void)
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_CGROUP_SOCKOPT,
 					       &bpf_mptcp_common_kfunc_set);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS,
-					       &bpf_mptcp_sched_kfunc_set);
+					       &bpf_mptcp_common_kfunc_set);
 #ifdef CONFIG_BPF_JIT
 	ret = ret ?: register_bpf_struct_ops(&bpf_mptcp_sched_ops, mptcp_sched_ops);
 #endif
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Use the newly added bpf_for_each() helper to walk the conn_list.
Drop bpf_mptcp_subflow_ctx_by_pos declaration.

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf.h b/tools/testing/selftests/bpf/progs/mptcp_bpf.h
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf.h
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf.h
@@ -XXX,XX +XXX,XX @@ bpf_mptcp_subflow_tcp_sock(const struct mptcp_subflow_context *subflow) __ksym;
 extern void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,
 					bool scheduled) __ksym;
 
-extern struct mptcp_subflow_context *
-bpf_mptcp_subflow_ctx_by_pos(const struct mptcp_sched_data *data, unsigned int pos) __ksym;
-
 #endif
diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_first.c
@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
 int BPF_PROG(bpf_first_get_send, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
-	mptcp_subflow_set_scheduled(bpf_mptcp_subflow_ctx_by_pos(data, 0), true);
+	struct mptcp_subflow_context *subflow;
+
+	subflow = bpf_mptcp_subflow_ctx(msk->first);
+	if (!subflow)
+		return -1;
+
+	mptcp_subflow_set_scheduled(subflow, true);
 	return 0;
 }
 
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Use the newly added bpf_for_each() helper to walk the conn_list.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../testing/selftests/bpf/progs/mptcp_bpf_bkup.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c
@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
 int BPF_PROG(bpf_bkup_get_send, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
-	int nr = -1;
-
-	for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
-		struct mptcp_subflow_context *subflow;
-
-		subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
-		if (!subflow)
-			break;
+	struct mptcp_subflow_context *subflow;
 
+	bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk) {
 		if (!BPF_CORE_READ_BITFIELD_PROBED(subflow, backup) ||
 		    !BPF_CORE_READ_BITFIELD_PROBED(subflow, request_bkup)) {
-			nr = i;
+			mptcp_subflow_set_scheduled(subflow, true);
 			break;
 		}
 	}
 
-	if (nr != -1) {
-		mptcp_subflow_set_scheduled(bpf_mptcp_subflow_ctx_by_pos(data, nr), true);
-		return -1;
-	}
 	return 0;
 }
 
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Use the newly added bpf_for_each() helper to walk the conn_list.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/bpf/progs/mptcp_bpf_rr.c        | 31 +++++++++----------
 1 file changed, 14 insertions(+), 17 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c
@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
 int BPF_PROG(bpf_rr_get_send, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
-	struct mptcp_subflow_context *subflow;
+	struct mptcp_subflow_context *subflow, *next;
 	struct mptcp_rr_storage *ptr;
-	struct sock *last_snd = NULL;
-	int nr = 0;
 
 	ptr = bpf_sk_storage_get(&mptcp_rr_map, msk, 0,
 				 BPF_LOCAL_STORAGE_GET_F_CREATE);
 	if (!ptr)
 		return -1;
 
-	last_snd = ptr->last_snd;
+	next = bpf_mptcp_subflow_ctx(msk->first);
+	if (!next)
+		return -1;
 
-	for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
-		subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
-		if (!last_snd || !subflow)
-			break;
+	if (!ptr->last_snd)
+		goto out;
 
-		if (mptcp_subflow_tcp_sock(subflow) == last_snd) {
-			if (i + 1 == MPTCP_SUBFLOWS_MAX ||
-			    !bpf_mptcp_subflow_ctx_by_pos(data, i + 1))
+	bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk) {
+		if (mptcp_subflow_tcp_sock(subflow) == ptr->last_snd) {
+			subflow = bpf_iter_mptcp_subflow_next(&___it);
+			if (!subflow)
 				break;
 
-			nr = i + 1;
+			next = subflow;
 			break;
 		}
 	}
 
-	subflow = bpf_mptcp_subflow_ctx_by_pos(data, nr);
-	if (!subflow)
-		return -1;
-	mptcp_subflow_set_scheduled(subflow, true);
-	ptr->last_snd = mptcp_subflow_tcp_sock(subflow);
+out:
+	mptcp_subflow_set_scheduled(next, true);
+	ptr->last_snd = mptcp_subflow_tcp_sock(next);
 	return 0;
 }
 
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Use the newly added bpf_for_each() helper to walk the conn_list.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/bpf/progs/mptcp_bpf_red.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
 int BPF_PROG(bpf_red_get_send, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
-	for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
-		if (!bpf_mptcp_subflow_ctx_by_pos(data, i))
-			break;
+	struct mptcp_subflow_context *subflow;
 
-		mptcp_subflow_set_scheduled(bpf_mptcp_subflow_ctx_by_pos(data, i), true);
-	}
+	bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk)
+		mptcp_subflow_set_scheduled(subflow, true);
 
 	return 0;
 }
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

Use the newly added bpf_for_each() helper to walk the conn_list.
Drop bpf_subflow_send_info, use subflow_send_info instead.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/bpf/progs/mptcp_bpf_burst.c     | 78 +++++++------------
 1 file changed, 26 insertions(+), 52 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c b/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c
index XXXXXXX..XXXXXXX 100644
--- a/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf_burst.c
@@ -XXX,XX +XXX,XX @@ char _license[] SEC("license") = "GPL";
 
 #define min(a, b) ((a) < (b) ? (a) : (b))
 
-struct bpf_subflow_send_info {
-	__u8 subflow_id;
-	__u64 linger_time;
-};
-
 extern bool mptcp_subflow_active(struct mptcp_subflow_context *subflow) __ksym;
 extern void mptcp_set_timeout(struct sock *sk) __ksym;
 extern __u64 mptcp_wnd_end(const struct mptcp_sock *msk) __ksym;
-extern bool tcp_stream_memory_free(const struct sock *sk, int wake) __ksym;
+extern bool bpf_sk_stream_memory_free(const struct sock *sk) __ksym;
 extern bool bpf_mptcp_subflow_queues_empty(struct sock *sk) __ksym;
 extern void mptcp_pm_subflow_chk_stale(const struct mptcp_sock *msk, struct sock *ssk) __ksym;
 
@@ -XXX,XX +XXX,XX @@ static __always_inline bool tcp_rtx_and_write_queues_empty(struct sock *sk)
 	return bpf_mptcp_subflow_queues_empty(sk) && tcp_write_queue_empty(sk);
 }
 
-static __always_inline bool __sk_stream_memory_free(const struct sock *sk, int wake)
-{
-	if (sk->sk_wmem_queued >= sk->sk_sndbuf)
-		return false;
-
-	return tcp_stream_memory_free(sk, wake);
-}
-
-static __always_inline bool sk_stream_memory_free(const struct sock *sk)
-{
-	return __sk_stream_memory_free(sk, 0);
-}
-
 SEC("struct_ops")
 void BPF_PROG(mptcp_sched_burst_init, struct mptcp_sock *msk)
 {
@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
 int BPF_PROG(bpf_burst_get_send, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
-	struct bpf_subflow_send_info send_info[SSK_MODE_MAX];
+	struct subflow_send_info send_info[SSK_MODE_MAX];
 	struct mptcp_subflow_context *subflow;
 	struct sock *sk = (struct sock *)msk;
 	__u32 pace, burst, wmem;
@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_burst_get_send, struct mptcp_sock *msk,
 
 	/* pick the subflow with the lower wmem/wspace ratio */
 	for (i = 0; i < SSK_MODE_MAX; ++i) {
-		send_info[i].subflow_id = MPTCP_SUBFLOWS_MAX;
+		send_info[i].ssk = NULL;
 		send_info[i].linger_time = -1;
 	}
 
-	for (i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
-		bool backup;
-
-		subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
-		if (!subflow)
-			break;
-
-		backup = subflow->backup || subflow->request_bkup;
+	bpf_for_each(mptcp_subflow, subflow, sk) {
+		bool backup = subflow->backup || subflow->request_bkup;
 
 		ssk = mptcp_subflow_tcp_sock(subflow);
 		if (!mptcp_subflow_active(subflow))
@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_burst_get_send, struct mptcp_sock *msk,
 
 		linger_time = div_u64((__u64)ssk->sk_wmem_queued << 32, pace);
 		if (linger_time < send_info[backup].linger_time) {
-			send_info[backup].subflow_id = i;
+			send_info[backup].ssk = ssk;
 			send_info[backup].linger_time = linger_time;
 		}
 	}
@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_burst_get_send, struct mptcp_sock *msk,
 
 	/* pick the best backup if no other subflow is active */
 	if (!nr_active)
-		send_info[SSK_MODE_ACTIVE].subflow_id = send_info[SSK_MODE_BACKUP].subflow_id;
+		send_info[SSK_MODE_ACTIVE].ssk = send_info[SSK_MODE_BACKUP].ssk;
 
-	subflow = bpf_mptcp_subflow_ctx_by_pos(data, send_info[SSK_MODE_ACTIVE].subflow_id);
-	if (!subflow)
+	ssk = send_info[SSK_MODE_ACTIVE].ssk;
+	if (!ssk || !bpf_sk_stream_memory_free(ssk))
 		return -1;
-	ssk = mptcp_subflow_tcp_sock(subflow);
-	if (!ssk || !sk_stream_memory_free(ssk))
+
+	subflow = bpf_mptcp_subflow_ctx(ssk);
+	if (!subflow)
 		return -1;
 
 	burst = min(MPTCP_SEND_BURST_SIZE, mptcp_wnd_end(msk) - msk->snd_nxt);
+	ssk = bpf_core_cast(ssk, struct sock);
 	wmem = ssk->sk_wmem_queued;
 	if (!burst)
 		goto out;
@@ -XXX,XX +XXX,XX @@ SEC("struct_ops")
 int BPF_PROG(bpf_burst_get_retrans, struct mptcp_sock *msk,
 	     struct mptcp_sched_data *data)
 {
-	int backup = MPTCP_SUBFLOWS_MAX, pick = MPTCP_SUBFLOWS_MAX, subflow_id;
+	struct sock *backup = NULL, *pick = NULL;
 	struct mptcp_subflow_context *subflow;
 	int min_stale_count = INT_MAX;
-	struct sock *ssk;
 
-	for (int i = 0; i < data->subflows && i < MPTCP_SUBFLOWS_MAX; i++) {
-		subflow = bpf_mptcp_subflow_ctx_by_pos(data, i);
-		if (!subflow)
-			break;
+	bpf_for_each(mptcp_subflow, subflow, (struct sock *)msk) {
+		struct sock *ssk = bpf_mptcp_subflow_tcp_sock(subflow);
 
-		if (!mptcp_subflow_active(subflow))
+		if (!ssk || !mptcp_subflow_active(subflow))
 			continue;
 
-		ssk = mptcp_subflow_tcp_sock(subflow);
 		/* still data outstanding at TCP level? skip this */
 		if (!tcp_rtx_and_write_queues_empty(ssk)) {
 			mptcp_pm_subflow_chk_stale(msk, ssk);
@@ -XXX,XX +XXX,XX @@ int BPF_PROG(bpf_burst_get_retrans, struct mptcp_sock *msk,
 		}
 
 		if (subflow->backup || subflow->request_bkup) {
-			if (backup == MPTCP_SUBFLOWS_MAX)
-				backup = i;
+			if (!backup)
+				backup = ssk;
 			continue;
 		}
 
-		if (pick == MPTCP_SUBFLOWS_MAX)
-			pick = i;
+		if (!pick)
+			pick = ssk;
 	}
 
-	if (pick < MPTCP_SUBFLOWS_MAX) {
-		subflow_id = pick;
+	if (pick)
 		goto out;
-	}
-	subflow_id = min_stale_count > 1 ? backup : MPTCP_SUBFLOWS_MAX;
+	pick = min_stale_count > 1 ? backup : NULL;
 
 out:
-	subflow = bpf_mptcp_subflow_ctx_by_pos(data, subflow_id);
+	if (!pick)
+		return -1;
+	subflow = bpf_mptcp_subflow_ctx(pick);
 	if (!subflow)
 		return -1;
 	mptcp_subflow_set_scheduled(subflow, true);
-- 
2.43.0

From: Geliang Tang <tanggeliang@kylinos.cn>

The mptcp_subflow bpf_iter is added now, it's better to use the helper
bpf_for_each(mptcp_subflow) to traverse all subflows on the conn_list of
an MPTCP socket and then call kfunc to modify the fields of each subflow
in the WIP MPTCP BPF packet scheduler examples, instead of converting them
to a fixed array. With this helper, we can get rid of this subflow array
"contexts" in struct mptcp_sched_data.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 include/net/mptcp.h | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index XXXXXXX..XXXXXXX 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -XXX,XX +XXX,XX @@ struct mptcp_out_options {
 #define MPTCP_SCHED_MAX		128
 #define MPTCP_SCHED_BUF_MAX	(MPTCP_SCHED_NAME_MAX * MPTCP_SCHED_MAX)
 
-#define MPTCP_SUBFLOWS_MAX	8
-
 struct mptcp_sched_data {
 	u8	subflows;
-	struct mptcp_subflow_context *contexts[MPTCP_SUBFLOWS_MAX];
 };
 
 struct mptcp_sched_ops {
-- 
2.43.0