From: Menglong Dong <imagedong@tencent.com>
Do the statistics of mptcp socket in use with sock_prot_inuse_add().
Therefore, we can get the count of used mptcp socket from
/proc/net/protocols:
& cat /proc/net/protocols
protocol size sockets memory press maxhdr slab module cl co di ac io in de sh ss gs se re sp bi br ha uh gp em
MPTCPv6 2048 0 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n
MPTCP 1896 1 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n
Signed-off-by: Menglong Dong <imagedong@tencent.com>
---
v8:
- remove the MPTCP_INUSE flag and do the statistics according to the
creation and destruction of the token
v6:
- introduce the 'MPTCP_INUSE' flag and check if msk is in use by it
v5:
- rebase to solve merge conflict
v4:
- rename MPTCP_DESTROIED to MPTCP_DESTROYED
v2:
- decrease the statistics for listening mptcp socket inuse with
mptcp_listen_inuse_dec()
- add MPTCP_DESTROIED flags to store if mptcp_destroy_common() was
called on the msk. For fallback case, we need to decrease the
---
net/mptcp/protocol.c | 16 +++++++++++++++-
net/mptcp/token.c | 7 +++++++
2 files changed, 22 insertions(+), 1 deletion(-)
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index ffdccf26af99..775454a9d49b 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -3005,6 +3005,16 @@ void mptcp_copy_inaddrs(struct sock *msk, const struct sock *ssk)
inet_sk(msk)->inet_rcv_saddr = inet_sk(ssk)->inet_rcv_saddr;
}
+static void mptcp_listen_inuse_dec(struct sock *sk)
+{
+ struct mptcp_sock *msk = mptcp_sk(sk);
+ struct socket *ssock;
+
+ ssock = __mptcp_nmpc_socket(msk);
+ if (ssock && inet_sk_state_load(ssock->sk) == TCP_LISTEN)
+ sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
+}
+
static int mptcp_disconnect(struct sock *sk, int flags)
{
struct mptcp_sock *msk = mptcp_sk(sk);
@@ -3025,6 +3035,7 @@ static int mptcp_disconnect(struct sock *sk, int flags)
if (msk->token)
mptcp_event(MPTCP_EVENT_CLOSED, msk, NULL, GFP_KERNEL);
+ mptcp_listen_inuse_dec(sk);
/* msk->subflow is still intact, the following will not free the first
* subflow
*/
@@ -3203,6 +3214,7 @@ static void mptcp_destroy(struct sock *sk)
{
struct mptcp_sock *msk = mptcp_sk(sk);
+ mptcp_listen_inuse_dec(sk);
/* clears msk->subflow, allowing the following to close
* even the initial subflow
*/
@@ -3675,8 +3687,10 @@ static int mptcp_listen(struct socket *sock, int backlog)
err = ssock->ops->listen(ssock, backlog);
inet_sk_state_store(sk, inet_sk_state_load(ssock->sk));
- if (!err)
+ if (!err) {
+ sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
mptcp_copy_inaddrs(sk, ssock->sk);
+ }
mptcp_event_pm_listener(ssock->sk, MPTCP_EVENT_LISTENER_CREATED);
diff --git a/net/mptcp/token.c b/net/mptcp/token.c
index 65430f314a68..54064146175f 100644
--- a/net/mptcp/token.c
+++ b/net/mptcp/token.c
@@ -175,6 +175,9 @@ int mptcp_token_new_connect(struct sock *sk)
__sk_nulls_add_node_rcu((struct sock *)msk, &bucket->msk_chain);
bucket->chain_len++;
spin_unlock_bh(&bucket->lock);
+ sock_prot_inuse_add(sock_net(subflow->conn),
+ subflow->conn->sk_prot,
+ 1);
return 0;
}
@@ -190,8 +193,10 @@ void mptcp_token_accept(struct mptcp_subflow_request_sock *req,
struct mptcp_sock *msk)
{
struct mptcp_subflow_request_sock *pos;
+ struct sock *sk = (struct sock *)msk;
struct token_bucket *bucket;
+ sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
bucket = token_bucket(req->token);
spin_lock_bh(&bucket->lock);
@@ -370,12 +375,14 @@ void mptcp_token_destroy_request(struct request_sock *req)
*/
void mptcp_token_destroy(struct mptcp_sock *msk)
{
+ struct sock *sk = (struct sock *)msk;
struct token_bucket *bucket;
struct mptcp_sock *pos;
if (sk_unhashed((struct sock *)msk))
return;
+ sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
bucket = token_bucket(msk->token);
spin_lock_bh(&bucket->lock);
pos = __token_lookup_msk(bucket, msk->token);
--
2.37.2On Thu, 2022-12-08 at 10:45 +0800, menglong8.dong@gmail.com wrote: > diff --git a/net/mptcp/token.c b/net/mptcp/token.c > index 65430f314a68..54064146175f 100644 > --- a/net/mptcp/token.c > +++ b/net/mptcp/token.c > @@ -175,6 +175,9 @@ int mptcp_token_new_connect(struct sock *sk) > __sk_nulls_add_node_rcu((struct sock *)msk, &bucket->msk_chain); > bucket->chain_len++; > spin_unlock_bh(&bucket->lock); > + sock_prot_inuse_add(sock_net(subflow->conn), > + subflow->conn->sk_prot, > + 1); Whoops, sorry for not noticing the above earlier... Here it would be better rename the mptcp_token_new_connect argument to 'ssk', and add a new local variable: struct sock *sk = subflow->conn; and replace the '(struct sock *)msk', 'subflow->conn' above with 'sk'. Side note: the kunit crash is due to build_icsk() being too naive: it creates a completely zeroed inet_connection_sock, while after this patch mptcp_token_new_connect() requires that the sk_prot field is initialized, too. The issue could be addresses with an addtional chunk in token_test.c. Note that for token's test sake, sk_prot could be initialized to any available/already exported proto. --- diff --git a/net/mptcp/token_test.c b/net/mptcp/token_test.c index 5d984bec1cd8..3bffe8d0d36f 100644 --- a/net/mptcp/token_test.c +++ b/net/mptcp/token_test.c @@ -36,6 +36,9 @@ static struct inet_connection_sock *build_icsk(struct kunit *test) icsk = kunit_kzalloc(test, sizeof(struct inet_connection_sock), GFP_USER); KUNIT_EXPECT_NOT_ERR_OR_NULL(test, icsk); + + /* be sure the tocken helpers can dereference sk->sk_prot */ + icsk->icsk_inet.sk.sk_prot = &tcp_prot; return icsk; }
On Tue, Dec 13, 2022 at 1:03 AM Paolo Abeni <pabeni@redhat.com> wrote: > > On Thu, 2022-12-08 at 10:45 +0800, menglong8.dong@gmail.com wrote: > > diff --git a/net/mptcp/token.c b/net/mptcp/token.c > > index 65430f314a68..54064146175f 100644 > > --- a/net/mptcp/token.c > > +++ b/net/mptcp/token.c > > @@ -175,6 +175,9 @@ int mptcp_token_new_connect(struct sock *sk) > > __sk_nulls_add_node_rcu((struct sock *)msk, &bucket->msk_chain); > > bucket->chain_len++; > > spin_unlock_bh(&bucket->lock); > > + sock_prot_inuse_add(sock_net(subflow->conn), > > + subflow->conn->sk_prot, > > + 1); > > Whoops, sorry for not noticing the above earlier... > Here it would be better rename the mptcp_token_new_connect argument to > 'ssk', and add a new local variable: > > struct sock *sk = subflow->conn; > > and replace the '(struct sock *)msk', 'subflow->conn' above with 'sk'. > Okay! > Side note: the kunit crash is due to build_icsk() being too naive: it > creates a completely zeroed inet_connection_sock, while after this > patch mptcp_token_new_connect() requires that the sk_prot field is > initialized, too. > > The issue could be addresses with an addtional chunk in token_test.c. > Note that for token's test sake, sk_prot could be initialized to any > available/already exported proto. > --- > diff --git a/net/mptcp/token_test.c b/net/mptcp/token_test.c > index 5d984bec1cd8..3bffe8d0d36f 100644 > --- a/net/mptcp/token_test.c > +++ b/net/mptcp/token_test.c > @@ -36,6 +36,9 @@ static struct inet_connection_sock *build_icsk(struct kunit *test) > icsk = kunit_kzalloc(test, sizeof(struct inet_connection_sock), > GFP_USER); > KUNIT_EXPECT_NOT_ERR_OR_NULL(test, icsk); > + > + /* be sure the tocken helpers can dereference sk->sk_prot */ > + icsk->icsk_inet.sk.sk_prot = &tcp_prot; > return icsk; > } > Get it! Thanks~ Menglong Dong >
On Thu, 2022-12-08 at 10:45 +0800, menglong8.dong@gmail.com wrote:
> From: Menglong Dong <imagedong@tencent.com>
>
> Do the statistics of mptcp socket in use with sock_prot_inuse_add().
> Therefore, we can get the count of used mptcp socket from
> /proc/net/protocols:
>
> & cat /proc/net/protocols
> protocol size sockets memory press maxhdr slab module cl co di ac io in de sh ss gs se re sp bi br ha uh gp em
> MPTCPv6 2048 0 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n
> MPTCP 1896 1 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n
>
> Signed-off-by: Menglong Dong <imagedong@tencent.com>
> ---
> v8:
> - remove the MPTCP_INUSE flag and do the statistics according to the
> creation and destruction of the token
>
> v6:
> - introduce the 'MPTCP_INUSE' flag and check if msk is in use by it
>
> v5:
> - rebase to solve merge conflict
>
> v4:
> - rename MPTCP_DESTROIED to MPTCP_DESTROYED
>
> v2:
> - decrease the statistics for listening mptcp socket inuse with
> mptcp_listen_inuse_dec()
> - add MPTCP_DESTROIED flags to store if mptcp_destroy_common() was
> called on the msk. For fallback case, we need to decrease the
> ---
> net/mptcp/protocol.c | 16 +++++++++++++++-
> net/mptcp/token.c | 7 +++++++
> 2 files changed, 22 insertions(+), 1 deletion(-)
>
> diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
> index ffdccf26af99..775454a9d49b 100644
> --- a/net/mptcp/protocol.c
> +++ b/net/mptcp/protocol.c
> @@ -3005,6 +3005,16 @@ void mptcp_copy_inaddrs(struct sock *msk, const struct sock *ssk)
> inet_sk(msk)->inet_rcv_saddr = inet_sk(ssk)->inet_rcv_saddr;
> }
>
> +static void mptcp_listen_inuse_dec(struct sock *sk)
> +{
> + struct mptcp_sock *msk = mptcp_sk(sk);
> + struct socket *ssock;
> +
> + ssock = __mptcp_nmpc_socket(msk);
> + if (ssock && inet_sk_state_load(ssock->sk) == TCP_LISTEN)
Here I think you can check directly: sk->sk_state == TCP_LISTEN, which
should be both simpler and safer, since this should be called under the
(m)sk socket lock.
It would be nice adding a comment describing why the state check is
needed, e.g.
/* Listener sockets do not have a valid token, but we need to
* account them, too.
*/
Other then that, LGTM.
Cheers,
Paolo
Hello,
On Tue, Dec 13, 2022 at 12:35 AM Paolo Abeni <pabeni@redhat.com> wrote:
>
> On Thu, 2022-12-08 at 10:45 +0800, menglong8.dong@gmail.com wrote:
> > From: Menglong Dong <imagedong@tencent.com>
> >
> > Do the statistics of mptcp socket in use with sock_prot_inuse_add().
> > Therefore, we can get the count of used mptcp socket from
> > /proc/net/protocols:
> >
> > & cat /proc/net/protocols
> > protocol size sockets memory press maxhdr slab module cl co di ac io in de sh ss gs se re sp bi br ha uh gp em
> > MPTCPv6 2048 0 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n
> > MPTCP 1896 1 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n
> >
> > Signed-off-by: Menglong Dong <imagedong@tencent.com>
> > ---
> > v8:
> > - remove the MPTCP_INUSE flag and do the statistics according to the
> > creation and destruction of the token
> >
> > v6:
> > - introduce the 'MPTCP_INUSE' flag and check if msk is in use by it
> >
> > v5:
> > - rebase to solve merge conflict
> >
> > v4:
> > - rename MPTCP_DESTROIED to MPTCP_DESTROYED
> >
> > v2:
> > - decrease the statistics for listening mptcp socket inuse with
> > mptcp_listen_inuse_dec()
> > - add MPTCP_DESTROIED flags to store if mptcp_destroy_common() was
> > called on the msk. For fallback case, we need to decrease the
> > ---
> > net/mptcp/protocol.c | 16 +++++++++++++++-
> > net/mptcp/token.c | 7 +++++++
> > 2 files changed, 22 insertions(+), 1 deletion(-)
> >
> > diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
> > index ffdccf26af99..775454a9d49b 100644
> > --- a/net/mptcp/protocol.c
> > +++ b/net/mptcp/protocol.c
> > @@ -3005,6 +3005,16 @@ void mptcp_copy_inaddrs(struct sock *msk, const struct sock *ssk)
> > inet_sk(msk)->inet_rcv_saddr = inet_sk(ssk)->inet_rcv_saddr;
> > }
> >
> > +static void mptcp_listen_inuse_dec(struct sock *sk)
> > +{
> > + struct mptcp_sock *msk = mptcp_sk(sk);
> > + struct socket *ssock;
> > +
> > + ssock = __mptcp_nmpc_socket(msk);
> > + if (ssock && inet_sk_state_load(ssock->sk) == TCP_LISTEN)
>
> Here I think you can check directly: sk->sk_state == TCP_LISTEN, which
> should be both simpler and safer, since this should be called under the
> (m)sk socket lock.
>
Then, mptcp_listen_inuse_dec() needs to be called before msk
enters TCP_CLOSE state, which means __mptcp_close()
and mptcp_disconnect(). Do I miss any case that a listening
msk enters TCP_CLOSE state?
> It would be nice adding a comment describing why the state check is
> needed, e.g.
>
> /* Listener sockets do not have a valid token, but we need to
> * account them, too.
> */
>
Sounds nice!
> Other then that, LGTM.
>
> Cheers,
>
> Paolo
>
On Fri, 2022-12-16 at 16:01 +0800, Menglong Dong wrote:
> Hello,
>
> On Tue, Dec 13, 2022 at 12:35 AM Paolo Abeni <pabeni@redhat.com> wrote:
> >
> > On Thu, 2022-12-08 at 10:45 +0800, menglong8.dong@gmail.com wrote:
> > > From: Menglong Dong <imagedong@tencent.com>
> > >
> > > Do the statistics of mptcp socket in use with sock_prot_inuse_add().
> > > Therefore, we can get the count of used mptcp socket from
> > > /proc/net/protocols:
> > >
> > > & cat /proc/net/protocols
> > > protocol size sockets memory press maxhdr slab module cl co di ac io in de sh ss gs se re sp bi br ha uh gp em
> > > MPTCPv6 2048 0 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n
> > > MPTCP 1896 1 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n
> > >
> > > Signed-off-by: Menglong Dong <imagedong@tencent.com>
> > > ---
> > > v8:
> > > - remove the MPTCP_INUSE flag and do the statistics according to the
> > > creation and destruction of the token
> > >
> > > v6:
> > > - introduce the 'MPTCP_INUSE' flag and check if msk is in use by it
> > >
> > > v5:
> > > - rebase to solve merge conflict
> > >
> > > v4:
> > > - rename MPTCP_DESTROIED to MPTCP_DESTROYED
> > >
> > > v2:
> > > - decrease the statistics for listening mptcp socket inuse with
> > > mptcp_listen_inuse_dec()
> > > - add MPTCP_DESTROIED flags to store if mptcp_destroy_common() was
> > > called on the msk. For fallback case, we need to decrease the
> > > ---
> > > net/mptcp/protocol.c | 16 +++++++++++++++-
> > > net/mptcp/token.c | 7 +++++++
> > > 2 files changed, 22 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
> > > index ffdccf26af99..775454a9d49b 100644
> > > --- a/net/mptcp/protocol.c
> > > +++ b/net/mptcp/protocol.c
> > > @@ -3005,6 +3005,16 @@ void mptcp_copy_inaddrs(struct sock *msk, const struct sock *ssk)
> > > inet_sk(msk)->inet_rcv_saddr = inet_sk(ssk)->inet_rcv_saddr;
> > > }
> > >
> > > +static void mptcp_listen_inuse_dec(struct sock *sk)
> > > +{
> > > + struct mptcp_sock *msk = mptcp_sk(sk);
> > > + struct socket *ssock;
> > > +
> > > + ssock = __mptcp_nmpc_socket(msk);
> > > + if (ssock && inet_sk_state_load(ssock->sk) == TCP_LISTEN)
> >
> > Here I think you can check directly: sk->sk_state == TCP_LISTEN, which
> > should be both simpler and safer, since this should be called under the
> > (m)sk socket lock.
> >
>
> Then, mptcp_listen_inuse_dec() needs to be called before msk
> enters TCP_CLOSE state, which means __mptcp_close()
> and mptcp_disconnect(). Do I miss any case that a listening
> msk enters TCP_CLOSE state?
AFAIK, that is correct. Note that the number of hooks will not change
compared to the current patch revision (mptcp_disconnect() +
mptcp_destroy()).
Cheers,
Paolo
Hello,
On Fri, Dec 16, 2022 at 4:45 PM Paolo Abeni <pabeni@redhat.com> wrote:
>
> On Fri, 2022-12-16 at 16:01 +0800, Menglong Dong wrote:
> > Hello,
> >
> > On Tue, Dec 13, 2022 at 12:35 AM Paolo Abeni <pabeni@redhat.com> wrote:
> > >
> > > On Thu, 2022-12-08 at 10:45 +0800, menglong8.dong@gmail.com wrote:
> > > > From: Menglong Dong <imagedong@tencent.com>
> > > >
> > > > Do the statistics of mptcp socket in use with sock_prot_inuse_add().
> > > > Therefore, we can get the count of used mptcp socket from
> > > > /proc/net/protocols:
> > > >
> > > > & cat /proc/net/protocols
> > > > protocol size sockets memory press maxhdr slab module cl co di ac io in de sh ss gs se re sp bi br ha uh gp em
> > > > MPTCPv6 2048 0 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n
> > > > MPTCP 1896 1 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n
> > > >
> > > > Signed-off-by: Menglong Dong <imagedong@tencent.com>
> > > > ---
> > > > v8:
> > > > - remove the MPTCP_INUSE flag and do the statistics according to the
> > > > creation and destruction of the token
> > > >
> > > > v6:
> > > > - introduce the 'MPTCP_INUSE' flag and check if msk is in use by it
> > > >
> > > > v5:
> > > > - rebase to solve merge conflict
> > > >
> > > > v4:
> > > > - rename MPTCP_DESTROIED to MPTCP_DESTROYED
> > > >
> > > > v2:
> > > > - decrease the statistics for listening mptcp socket inuse with
> > > > mptcp_listen_inuse_dec()
> > > > - add MPTCP_DESTROIED flags to store if mptcp_destroy_common() was
> > > > called on the msk. For fallback case, we need to decrease the
> > > > ---
> > > > net/mptcp/protocol.c | 16 +++++++++++++++-
> > > > net/mptcp/token.c | 7 +++++++
> > > > 2 files changed, 22 insertions(+), 1 deletion(-)
> > > >
> > > > diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
> > > > index ffdccf26af99..775454a9d49b 100644
> > > > --- a/net/mptcp/protocol.c
> > > > +++ b/net/mptcp/protocol.c
> > > > @@ -3005,6 +3005,16 @@ void mptcp_copy_inaddrs(struct sock *msk, const struct sock *ssk)
> > > > inet_sk(msk)->inet_rcv_saddr = inet_sk(ssk)->inet_rcv_saddr;
> > > > }
> > > >
> > > > +static void mptcp_listen_inuse_dec(struct sock *sk)
> > > > +{
> > > > + struct mptcp_sock *msk = mptcp_sk(sk);
> > > > + struct socket *ssock;
> > > > +
> > > > + ssock = __mptcp_nmpc_socket(msk);
> > > > + if (ssock && inet_sk_state_load(ssock->sk) == TCP_LISTEN)
> > >
> > > Here I think you can check directly: sk->sk_state == TCP_LISTEN, which
> > > should be both simpler and safer, since this should be called under the
> > > (m)sk socket lock.
> > >
> >
> > Then, mptcp_listen_inuse_dec() needs to be called before msk
> > enters TCP_CLOSE state, which means __mptcp_close()
> > and mptcp_disconnect(). Do I miss any case that a listening
> > msk enters TCP_CLOSE state?
>
> AFAIK, that is correct. Note that the number of hooks will not change
> compared to the current patch revision (mptcp_disconnect() +
> mptcp_destroy()).
>
With the changes you suggested, I need to the move mptcp_listen_inuse_dec()
that called in mptcp_destroy() to __mptcp_close(), as msk is already in
CLOSE state in mptcp_destroy(). Isn't it?
Thanks!
Menglong Dong
>
> Cheers,
>
> Paolo
>
Hi Menglong Dong, On 08/12/2022 03:45, menglong8.dong@gmail.com wrote: > From: Menglong Dong <imagedong@tencent.com> > > Do the statistics of mptcp socket in use with sock_prot_inuse_add(). > Therefore, we can get the count of used mptcp socket from > /proc/net/protocols: > > & cat /proc/net/protocols > protocol size sockets memory press maxhdr slab module cl co di ac io in de sh ss gs se re sp bi br ha uh gp em > MPTCPv6 2048 0 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n > MPTCP 1896 1 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n > > Signed-off-by: Menglong Dong <imagedong@tencent.com> > --- > v8: > - remove the MPTCP_INUSE flag and do the statistics according to the > creation and destruction of the token Thank you for the new version! I marked this series as "Changes Requested" on Patchwork because I guess you saw the CI was not happy with the KUnit token tests, e.g.: - KVM Validation: normal (except selftest_mptcp_join): - Unstable: 1 failed test(s): insmod - Critical: 3 Call Trace(s) ❌: - Task: https://cirrus-ci.com/task/5270512411082752 - Summary: https://api.cirrus-ci.com/v1/artifact/task/5270512411082752/summary/summary.txt - Debug: https://cirrus-ci.com/task/4989037434372096 Do you mind looking at that please? Cheers, Matt -- Tessares | Belgium | Hybrid Access Solutions www.tessares.net
On Thu, Dec 8, 2022 at 9:53 PM Matthieu Baerts <matthieu.baerts@tessares.net> wrote: > > Hi Menglong Dong, > > On 08/12/2022 03:45, menglong8.dong@gmail.com wrote: > > From: Menglong Dong <imagedong@tencent.com> > > > > Do the statistics of mptcp socket in use with sock_prot_inuse_add(). > > Therefore, we can get the count of used mptcp socket from > > /proc/net/protocols: > > > > & cat /proc/net/protocols > > protocol size sockets memory press maxhdr slab module cl co di ac io in de sh ss gs se re sp bi br ha uh gp em > > MPTCPv6 2048 0 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n > > MPTCP 1896 1 0 no 0 yes kernel y n y y y y y y y y y y n n n y y y n > > > > Signed-off-by: Menglong Dong <imagedong@tencent.com> > > --- > > v8: > > - remove the MPTCP_INUSE flag and do the statistics according to the > > creation and destruction of the token > > Thank you for the new version! > > I marked this series as "Changes Requested" on Patchwork because I guess > you saw the CI was not happy with the KUnit token tests, e.g.: > > - KVM Validation: normal (except selftest_mptcp_join): > - Unstable: 1 failed test(s): insmod - Critical: 3 Call Trace(s) ❌: > - Task: https://cirrus-ci.com/task/5270512411082752 > - Summary: > https://api.cirrus-ci.com/v1/artifact/task/5270512411082752/summary/summary.txt > - Debug: https://cirrus-ci.com/task/4989037434372096 > > Do you mind looking at that please? > Of course! I'll run all of the tests to make sure they pass. > Cheers, > Matt > -- > Tessares | Belgium | Hybrid Access Solutions > www.tessares.net
© 2016 - 2026 Red Hat, Inc.