* [PATCH net 0/6] mptcp: Connection and accounting fixes
@ 2021-07-10 0:20 Mat Martineau
2021-07-10 0:20 ` [PATCH net 1/6] mptcp: fix warning in __skb_flow_dissect() when do syn cookie for subflow join Mat Martineau
` (6 more replies)
0 siblings, 7 replies; 8+ messages in thread
From: Mat Martineau @ 2021-07-10 0:20 UTC (permalink / raw)
To: netdev
Cc: Mat Martineau, davem, kuba, matthieu.baerts, pabeni, fw,
geliangtang, mptcp
Here are some miscellaneous fixes for MPTCP:
Patch 1 modifies an MPTCP hash so it doesn't depend on one of skb->dev
and skb->sk being non-NULL.
Patch 2 removes an extra destructor call when rejecting a join due to
port mismatch.
Patches 3 and 5 more cleanly handle error conditions with MP_JOIN and
syncookies, and update a related self test.
Patch 4 makes sure packets that trigger a subflow TCP reset during MPTCP
option header processing are correctly dropped.
Patch 6 addresses a rmem accounting issue that could keep packets in
subflow receive buffers longer than necessary, delaying MPTCP-level
ACKs.
Jianguo Wu (5):
mptcp: fix warning in __skb_flow_dissect() when do syn cookie for
subflow join
mptcp: remove redundant req destruct in subflow_check_req()
mptcp: fix syncookie process if mptcp can not_accept new subflow
mptcp: avoid processing packet if a subflow reset
selftests: mptcp: fix case multiple subflows limited by server
Paolo Abeni (1):
mptcp: properly account bulk freed memory
include/net/mptcp.h | 5 +++--
net/ipv4/tcp_input.c | 19 +++++++++++++++----
net/mptcp/mib.c | 1 +
net/mptcp/mib.h | 1 +
net/mptcp/options.c | 19 +++++++++++++------
net/mptcp/protocol.c | 12 +++++++-----
net/mptcp/protocol.h | 10 +++++++++-
net/mptcp/subflow.c | 11 +++--------
net/mptcp/syncookies.c | 16 +++++++++++++++-
.../testing/selftests/net/mptcp/mptcp_join.sh | 2 +-
10 files changed, 68 insertions(+), 28 deletions(-)
--
2.32.0
^ permalink raw reply [flat|nested] 8+ messages in thread
* [PATCH net 1/6] mptcp: fix warning in __skb_flow_dissect() when do syn cookie for subflow join
2021-07-10 0:20 [PATCH net 0/6] mptcp: Connection and accounting fixes Mat Martineau
@ 2021-07-10 0:20 ` Mat Martineau
2021-07-10 0:20 ` [PATCH net 2/6] mptcp: remove redundant req destruct in subflow_check_req() Mat Martineau
` (5 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-07-10 0:20 UTC (permalink / raw)
To: netdev
Cc: Jianguo Wu, davem, kuba, matthieu.baerts, pabeni, fw, mptcp,
Mat Martineau
From: Jianguo Wu <wujianguo@chinatelecom.cn>
I did stress test with wrk[1] and webfsd[2] with the assistance of
mptcp-tools[3]:
Server side:
./use_mptcp.sh webfsd -4 -R /tmp/ -p 8099
Client side:
./use_mptcp.sh wrk -c 200 -d 30 -t 4 http://192.168.174.129:8099/
and got the following warning message:
[ 55.552626] TCP: request_sock_subflow: Possible SYN flooding on port 8099. Sending cookies. Check SNMP counters.
[ 55.553024] ------------[ cut here ]------------
[ 55.553027] WARNING: CPU: 0 PID: 10 at net/core/flow_dissector.c:984 __skb_flow_dissect+0x280/0x1650
...
[ 55.553117] CPU: 0 PID: 10 Comm: ksoftirqd/0 Not tainted 5.12.0+ #18
[ 55.553121] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 02/27/2020
[ 55.553124] RIP: 0010:__skb_flow_dissect+0x280/0x1650
...
[ 55.553133] RSP: 0018:ffffb79580087770 EFLAGS: 00010246
[ 55.553137] RAX: 0000000000000000 RBX: ffffffff8ddb58e0 RCX: ffffb79580087888
[ 55.553139] RDX: ffffffff8ddb58e0 RSI: ffff8f7e4652b600 RDI: 0000000000000000
[ 55.553141] RBP: ffffb79580087858 R08: 0000000000000000 R09: 0000000000000008
[ 55.553143] R10: 000000008c622965 R11: 00000000d3313a5b R12: ffff8f7e4652b600
[ 55.553146] R13: ffff8f7e465c9062 R14: 0000000000000000 R15: ffffb79580087888
[ 55.553149] FS: 0000000000000000(0000) GS:ffff8f7f75e00000(0000) knlGS:0000000000000000
[ 55.553152] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 55.553154] CR2: 00007f73d1d19000 CR3: 0000000135e10004 CR4: 00000000003706f0
[ 55.553160] Call Trace:
[ 55.553166] ? __sha256_final+0x67/0xd0
[ 55.553173] ? sha256+0x7e/0xa0
[ 55.553177] __skb_get_hash+0x57/0x210
[ 55.553182] subflow_init_req_cookie_join_save+0xac/0xc0
[ 55.553189] subflow_check_req+0x474/0x550
[ 55.553195] ? ip_route_output_key_hash+0x67/0x90
[ 55.553200] ? xfrm_lookup_route+0x1d/0xa0
[ 55.553207] subflow_v4_route_req+0x8e/0xd0
[ 55.553212] tcp_conn_request+0x31e/0xab0
[ 55.553218] ? selinux_socket_sock_rcv_skb+0x116/0x210
[ 55.553224] ? tcp_rcv_state_process+0x179/0x6d0
[ 55.553229] tcp_rcv_state_process+0x179/0x6d0
[ 55.553235] tcp_v4_do_rcv+0xaf/0x220
[ 55.553239] tcp_v4_rcv+0xce4/0xd80
[ 55.553243] ? ip_route_input_rcu+0x246/0x260
[ 55.553248] ip_protocol_deliver_rcu+0x35/0x1b0
[ 55.553253] ip_local_deliver_finish+0x44/0x50
[ 55.553258] ip_local_deliver+0x6c/0x110
[ 55.553262] ? ip_rcv_finish_core.isra.19+0x5a/0x400
[ 55.553267] ip_rcv+0xd1/0xe0
...
After debugging, I found in __skb_flow_dissect(), skb->dev and skb->sk
are both NULL, then net is NULL, and trigger WARN_ON_ONCE(!net),
actually net is always NULL in this code path, as skb->dev is set to
NULL in tcp_v4_rcv(), and skb->sk is never set.
Code snippet in __skb_flow_dissect() that trigger warning:
975 if (skb) {
976 if (!net) {
977 if (skb->dev)
978 net = dev_net(skb->dev);
979 else if (skb->sk)
980 net = sock_net(skb->sk);
981 }
982 }
983
984 WARN_ON_ONCE(!net);
So, using seq and transport header derived hash.
[1] https://github.com/wg/wrk
[2] https://github.com/ourway/webfsd
[3] https://github.com/pabeni/mptcp-tools
Fixes: 9466a1ccebbe ("mptcp: enable JOIN requests even if cookies are in use")
Suggested-by: Paolo Abeni <pabeni@redhat.com>
Suggested-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Jianguo Wu <wujianguo@chinatelecom.cn>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
net/mptcp/syncookies.c | 16 +++++++++++++++-
1 file changed, 15 insertions(+), 1 deletion(-)
diff --git a/net/mptcp/syncookies.c b/net/mptcp/syncookies.c
index abe0fd099746..37127781aee9 100644
--- a/net/mptcp/syncookies.c
+++ b/net/mptcp/syncookies.c
@@ -37,7 +37,21 @@ static spinlock_t join_entry_locks[COOKIE_JOIN_SLOTS] __cacheline_aligned_in_smp
static u32 mptcp_join_entry_hash(struct sk_buff *skb, struct net *net)
{
- u32 i = skb_get_hash(skb) ^ net_hash_mix(net);
+ static u32 mptcp_join_hash_secret __read_mostly;
+ struct tcphdr *th = tcp_hdr(skb);
+ u32 seq, i;
+
+ net_get_random_once(&mptcp_join_hash_secret,
+ sizeof(mptcp_join_hash_secret));
+
+ if (th->syn)
+ seq = TCP_SKB_CB(skb)->seq;
+ else
+ seq = TCP_SKB_CB(skb)->seq - 1;
+
+ i = jhash_3words(seq, net_hash_mix(net),
+ (__force __u32)th->source << 16 | (__force __u32)th->dest,
+ mptcp_join_hash_secret);
return i % ARRAY_SIZE(join_entries);
}
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net 2/6] mptcp: remove redundant req destruct in subflow_check_req()
2021-07-10 0:20 [PATCH net 0/6] mptcp: Connection and accounting fixes Mat Martineau
2021-07-10 0:20 ` [PATCH net 1/6] mptcp: fix warning in __skb_flow_dissect() when do syn cookie for subflow join Mat Martineau
@ 2021-07-10 0:20 ` Mat Martineau
2021-07-10 0:20 ` [PATCH net 3/6] mptcp: fix syncookie process if mptcp can not_accept new subflow Mat Martineau
` (4 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-07-10 0:20 UTC (permalink / raw)
To: netdev
Cc: Jianguo Wu, davem, kuba, matthieu.baerts, geliangtang, mptcp,
Mat Martineau
From: Jianguo Wu <wujianguo@chinatelecom.cn>
In subflow_check_req(), if subflow sport is mismatch, will put msk,
destroy token, and destruct req, then return -EPERM, which can be
done by subflow_req_destructor() via:
tcp_conn_request()
|--__reqsk_free()
|--subflow_req_destructor()
So we should remove these redundant code, otherwise will call
tcp_v4_reqsk_destructor() twice, and may double free
inet_rsk(req)->ireq_opt.
Fixes: 5bc56388c74f ("mptcp: add port number check for MP_JOIN")
Signed-off-by: Jianguo Wu <wujianguo@chinatelecom.cn>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
net/mptcp/subflow.c | 5 -----
1 file changed, 5 deletions(-)
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 66d0b1893d26..b15e2017168d 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -214,11 +214,6 @@ static int subflow_check_req(struct request_sock *req,
ntohs(inet_sk(sk_listener)->inet_sport),
ntohs(inet_sk((struct sock *)subflow_req->msk)->inet_sport));
if (!mptcp_pm_sport_in_anno_list(subflow_req->msk, sk_listener)) {
- sock_put((struct sock *)subflow_req->msk);
- mptcp_token_destroy_request(req);
- tcp_request_sock_ops.destructor(req);
- subflow_req->msk = NULL;
- subflow_req->mp_join = 0;
SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_MISMATCHPORTSYNRX);
return -EPERM;
}
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net 3/6] mptcp: fix syncookie process if mptcp can not_accept new subflow
2021-07-10 0:20 [PATCH net 0/6] mptcp: Connection and accounting fixes Mat Martineau
2021-07-10 0:20 ` [PATCH net 1/6] mptcp: fix warning in __skb_flow_dissect() when do syn cookie for subflow join Mat Martineau
2021-07-10 0:20 ` [PATCH net 2/6] mptcp: remove redundant req destruct in subflow_check_req() Mat Martineau
@ 2021-07-10 0:20 ` Mat Martineau
2021-07-10 0:20 ` [PATCH net 4/6] mptcp: avoid processing packet if a subflow reset Mat Martineau
` (3 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-07-10 0:20 UTC (permalink / raw)
To: netdev
Cc: Jianguo Wu, davem, kuba, matthieu.baerts, pabeni, fw, mptcp,
Mat Martineau
From: Jianguo Wu <wujianguo@chinatelecom.cn>
Lots of "TCP: tcp_fin: Impossible, sk->sk_state=7" in client side
when doing stress testing using wrk and webfsd.
There are at least two cases may trigger this warning:
1.mptcp is in syncookie, and server recv MP_JOIN SYN request,
in subflow_check_req(), the mptcp_can_accept_new_subflow()
return false, so subflow_init_req_cookie_join_save() isn't
called, i.e. not store the data present in the MP_JOIN syn
request and the random nonce in hash table - join_entries[],
but still send synack. When recv 3rd-ack,
mptcp_token_join_cookie_init_state() will return false, and
3rd-ack is dropped, then if mptcp conn is closed by client,
client will send a DATA_FIN and a MPTCP FIN, the DATA_FIN
doesn't have MP_CAPABLE or MP_JOIN,
so mptcp_subflow_init_cookie_req() will return 0, and pass
the cookie check, MP_JOIN request is fallback to normal TCP.
Server will send a TCP FIN if closed, in client side,
when process TCP FIN, it will do reset, the code path is:
tcp_data_queue()->mptcp_incoming_options()
->check_fully_established()->mptcp_subflow_reset().
mptcp_subflow_reset() will set sock state to TCP_CLOSE,
so tcp_fin will hit TCP_CLOSE, and print the warning.
2.mptcp is in syncookie, and server recv 3rd-ack, in
mptcp_subflow_init_cookie_req(), mptcp_can_accept_new_subflow()
return false, and subflow_req->mp_join is not set to 1,
so in subflow_syn_recv_sock() will not reset the MP_JOIN
subflow, but fallback to normal TCP, and then the same thing
happens when server will send a TCP FIN if closed.
For case1, subflow_check_req() return -EPERM,
then tcp_conn_request() will drop MP_JOIN SYN.
For case2, let subflow_syn_recv_sock() call
mptcp_can_accept_new_subflow(), and do fatal fallback, send reset.
Fixes: 9466a1ccebbe ("mptcp: enable JOIN requests even if cookies are in use")
Signed-off-by: Jianguo Wu <wujianguo@chinatelecom.cn>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
net/mptcp/subflow.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index b15e2017168d..966f777d35ce 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -225,6 +225,8 @@ static int subflow_check_req(struct request_sock *req,
if (unlikely(req->syncookie)) {
if (mptcp_can_accept_new_subflow(subflow_req->msk))
subflow_init_req_cookie_join_save(subflow_req, skb);
+ else
+ return -EPERM;
}
pr_debug("token=%u, remote_nonce=%u msk=%p", subflow_req->token,
@@ -264,9 +266,7 @@ int mptcp_subflow_init_cookie_req(struct request_sock *req,
if (!mptcp_token_join_cookie_init_state(subflow_req, skb))
return -EINVAL;
- if (mptcp_can_accept_new_subflow(subflow_req->msk))
- subflow_req->mp_join = 1;
-
+ subflow_req->mp_join = 1;
subflow_req->ssn_offset = TCP_SKB_CB(skb)->seq - 1;
}
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net 4/6] mptcp: avoid processing packet if a subflow reset
2021-07-10 0:20 [PATCH net 0/6] mptcp: Connection and accounting fixes Mat Martineau
` (2 preceding siblings ...)
2021-07-10 0:20 ` [PATCH net 3/6] mptcp: fix syncookie process if mptcp can not_accept new subflow Mat Martineau
@ 2021-07-10 0:20 ` Mat Martineau
2021-07-10 0:20 ` [PATCH net 5/6] selftests: mptcp: fix case multiple subflows limited by server Mat Martineau
` (2 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-07-10 0:20 UTC (permalink / raw)
To: netdev
Cc: Jianguo Wu, davem, kuba, matthieu.baerts, pabeni, mptcp, Mat Martineau
From: Jianguo Wu <wujianguo@chinatelecom.cn>
If check_fully_established() causes a subflow reset, it should not
continue to process the packet in tcp_data_queue().
Add a return value to mptcp_incoming_options(), and return false if a
subflow has been reset, else return true. Then drop the packet in
tcp_data_queue()/tcp_rcv_state_process() if mptcp_incoming_options()
return false.
Fixes: d582484726c4 ("mptcp: fix fallback for MP_JOIN subflows")
Signed-off-by: Jianguo Wu <wujianguo@chinatelecom.cn>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
include/net/mptcp.h | 5 +++--
net/ipv4/tcp_input.c | 19 +++++++++++++++----
net/mptcp/options.c | 19 +++++++++++++------
3 files changed, 31 insertions(+), 12 deletions(-)
diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index cb580b06152f..8b5af683a818 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -105,7 +105,7 @@ bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
unsigned int *size, unsigned int remaining,
struct mptcp_out_options *opts);
-void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb);
+bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb);
void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
struct mptcp_out_options *opts);
@@ -227,9 +227,10 @@ static inline bool mptcp_established_options(struct sock *sk,
return false;
}
-static inline void mptcp_incoming_options(struct sock *sk,
+static inline bool mptcp_incoming_options(struct sock *sk,
struct sk_buff *skb)
{
+ return true;
}
static inline void mptcp_skb_ext_move(struct sk_buff *to,
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a5a8d0a378b2..149ceb5c94ff 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4247,6 +4247,9 @@ void tcp_reset(struct sock *sk, struct sk_buff *skb)
{
trace_tcp_receive_reset(sk);
+ /* mptcp can't tell us to ignore reset pkts,
+ * so just ignore the return value of mptcp_incoming_options().
+ */
if (sk_is_mptcp(sk))
mptcp_incoming_options(sk, skb);
@@ -4941,8 +4944,13 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
bool fragstolen;
int eaten;
- if (sk_is_mptcp(sk))
- mptcp_incoming_options(sk, skb);
+ /* If a subflow has been reset, the packet should not continue
+ * to be processed, drop the packet.
+ */
+ if (sk_is_mptcp(sk) && !mptcp_incoming_options(sk, skb)) {
+ __kfree_skb(skb);
+ return;
+ }
if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) {
__kfree_skb(skb);
@@ -6523,8 +6531,11 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb)
case TCP_CLOSING:
case TCP_LAST_ACK:
if (!before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
- if (sk_is_mptcp(sk))
- mptcp_incoming_options(sk, skb);
+ /* If a subflow has been reset, the packet should not
+ * continue to be processed, drop the packet.
+ */
+ if (sk_is_mptcp(sk) && !mptcp_incoming_options(sk, skb))
+ goto discard;
break;
}
fallthrough;
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index b5850afea343..4452455aef7f 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -1035,7 +1035,8 @@ static bool add_addr_hmac_valid(struct mptcp_sock *msk,
return hmac == mp_opt->ahmac;
}
-void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb)
+/* Return false if a subflow has been reset, else return true */
+bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb)
{
struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
struct mptcp_sock *msk = mptcp_sk(subflow->conn);
@@ -1053,12 +1054,16 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb)
__mptcp_check_push(subflow->conn, sk);
__mptcp_data_acked(subflow->conn);
mptcp_data_unlock(subflow->conn);
- return;
+ return true;
}
mptcp_get_options(sk, skb, &mp_opt);
+
+ /* The subflow can be in close state only if check_fully_established()
+ * just sent a reset. If so, tell the caller to ignore the current packet.
+ */
if (!check_fully_established(msk, sk, subflow, skb, &mp_opt))
- return;
+ return sk->sk_state != TCP_CLOSE;
if (mp_opt.fastclose &&
msk->local_key == mp_opt.rcvr_key) {
@@ -1100,7 +1105,7 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb)
}
if (!mp_opt.dss)
- return;
+ return true;
/* we can't wait for recvmsg() to update the ack_seq, otherwise
* monodirectional flows will stuck
@@ -1119,12 +1124,12 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb)
schedule_work(&msk->work))
sock_hold(subflow->conn);
- return;
+ return true;
}
mpext = skb_ext_add(skb, SKB_EXT_MPTCP);
if (!mpext)
- return;
+ return true;
memset(mpext, 0, sizeof(*mpext));
@@ -1153,6 +1158,8 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb)
if (mpext->csum_reqd)
mpext->csum = mp_opt.csum;
}
+
+ return true;
}
static void mptcp_set_rwin(const struct tcp_sock *tp)
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net 5/6] selftests: mptcp: fix case multiple subflows limited by server
2021-07-10 0:20 [PATCH net 0/6] mptcp: Connection and accounting fixes Mat Martineau
` (3 preceding siblings ...)
2021-07-10 0:20 ` [PATCH net 4/6] mptcp: avoid processing packet if a subflow reset Mat Martineau
@ 2021-07-10 0:20 ` Mat Martineau
2021-07-10 0:20 ` [PATCH net 6/6] mptcp: properly account bulk freed memory Mat Martineau
2021-07-10 1:50 ` [PATCH net 0/6] mptcp: Connection and accounting fixes patchwork-bot+netdevbpf
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-07-10 0:20 UTC (permalink / raw)
To: netdev
Cc: Jianguo Wu, davem, kuba, matthieu.baerts, pabeni, fw, mptcp,
kernel test robot, Mat Martineau
From: Jianguo Wu <wujianguo@chinatelecom.cn>
After patch "mptcp: fix syncookie process if mptcp can not_accept new
subflow", if subflow is limited, MP_JOIN SYN is dropped, and no SYN/ACK
will be replied.
So in case "multiple subflows limited by server", the expected SYN/ACK
number should be 1.
Fixes: 00587187ad30 ("selftests: mptcp: add test cases for mptcp join tests with syn cookies")
Reported-by: kernel test robot <oliver.sang@intel.com>
Signed-off-by: Jianguo Wu <wujianguo@chinatelecom.cn>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
tools/testing/selftests/net/mptcp/mptcp_join.sh | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index 9a191c1a5de8..f02f4de2f3a0 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -1409,7 +1409,7 @@ syncookies_tests()
ip netns exec $ns2 ./pm_nl_ctl add 10.0.3.2 flags subflow
ip netns exec $ns2 ./pm_nl_ctl add 10.0.2.2 flags subflow
run_tests $ns1 $ns2 10.0.1.1
- chk_join_nr "subflows limited by server w cookies" 2 2 1
+ chk_join_nr "subflows limited by server w cookies" 2 1 1
# test signal address with cookies
reset_with_cookies
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net 6/6] mptcp: properly account bulk freed memory
2021-07-10 0:20 [PATCH net 0/6] mptcp: Connection and accounting fixes Mat Martineau
` (4 preceding siblings ...)
2021-07-10 0:20 ` [PATCH net 5/6] selftests: mptcp: fix case multiple subflows limited by server Mat Martineau
@ 2021-07-10 0:20 ` Mat Martineau
2021-07-10 1:50 ` [PATCH net 0/6] mptcp: Connection and accounting fixes patchwork-bot+netdevbpf
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-07-10 0:20 UTC (permalink / raw)
To: netdev
Cc: Paolo Abeni, davem, kuba, matthieu.baerts, fw, mptcp, Mat Martineau
From: Paolo Abeni <pabeni@redhat.com>
After commit 879526030c8b ("mptcp: protect the rx path with
the msk socket spinlock") the rmem currently used by a given
msk is really sk_rmem_alloc - rmem_released.
The safety check in mptcp_data_ready() does not take the above
in due account, as a result legit incoming data is kept in
subflow receive queue with no reason, delaying or blocking
MPTCP-level ack generation.
This change addresses the issue introducing a new helper to fetch
the rmem memory and using it as needed. Additionally add a MIB
counter for the exceptional event described above - the peer is
misbehaving.
Finally, introduce the required annotation when rmem_released is
updated.
Fixes: 879526030c8b ("mptcp: protect the rx path with the msk socket spinlock")
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/211
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
net/mptcp/mib.c | 1 +
net/mptcp/mib.h | 1 +
net/mptcp/protocol.c | 12 +++++++-----
net/mptcp/protocol.h | 10 +++++++++-
4 files changed, 18 insertions(+), 6 deletions(-)
diff --git a/net/mptcp/mib.c b/net/mptcp/mib.c
index 52ea2517e856..ff2cc0e3273d 100644
--- a/net/mptcp/mib.c
+++ b/net/mptcp/mib.c
@@ -44,6 +44,7 @@ static const struct snmp_mib mptcp_snmp_list[] = {
SNMP_MIB_ITEM("RmSubflow", MPTCP_MIB_RMSUBFLOW),
SNMP_MIB_ITEM("MPPrioTx", MPTCP_MIB_MPPRIOTX),
SNMP_MIB_ITEM("MPPrioRx", MPTCP_MIB_MPPRIORX),
+ SNMP_MIB_ITEM("RcvPruned", MPTCP_MIB_RCVPRUNED),
SNMP_MIB_SENTINEL
};
diff --git a/net/mptcp/mib.h b/net/mptcp/mib.h
index 193466c9b549..0663cb12b448 100644
--- a/net/mptcp/mib.h
+++ b/net/mptcp/mib.h
@@ -37,6 +37,7 @@ enum linux_mptcp_mib_field {
MPTCP_MIB_RMSUBFLOW, /* Remove a subflow */
MPTCP_MIB_MPPRIOTX, /* Transmit a MP_PRIO */
MPTCP_MIB_MPPRIORX, /* Received a MP_PRIO */
+ MPTCP_MIB_RCVPRUNED, /* Incoming packet dropped due to memory limit */
__MPTCP_MIB_MAX
};
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 7a5afa8c6866..a88924947815 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -474,7 +474,7 @@ static void mptcp_cleanup_rbuf(struct mptcp_sock *msk)
bool cleanup, rx_empty;
cleanup = (space > 0) && (space >= (old_space << 1));
- rx_empty = !atomic_read(&sk->sk_rmem_alloc);
+ rx_empty = !__mptcp_rmem(sk);
mptcp_for_each_subflow(msk, subflow) {
struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
@@ -720,8 +720,10 @@ void mptcp_data_ready(struct sock *sk, struct sock *ssk)
sk_rbuf = ssk_rbuf;
/* over limit? can't append more skbs to msk, Also, no need to wake-up*/
- if (atomic_read(&sk->sk_rmem_alloc) > sk_rbuf)
+ if (__mptcp_rmem(sk) > sk_rbuf) {
+ MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_RCVPRUNED);
return;
+ }
/* Wake-up the reader only for in-sequence data */
mptcp_data_lock(sk);
@@ -1754,7 +1756,7 @@ static int __mptcp_recvmsg_mskq(struct mptcp_sock *msk,
if (!(flags & MSG_PEEK)) {
/* we will bulk release the skb memory later */
skb->destructor = NULL;
- msk->rmem_released += skb->truesize;
+ WRITE_ONCE(msk->rmem_released, msk->rmem_released + skb->truesize);
__skb_unlink(skb, &msk->receive_queue);
__kfree_skb(skb);
}
@@ -1873,7 +1875,7 @@ static void __mptcp_update_rmem(struct sock *sk)
atomic_sub(msk->rmem_released, &sk->sk_rmem_alloc);
sk_mem_uncharge(sk, msk->rmem_released);
- msk->rmem_released = 0;
+ WRITE_ONCE(msk->rmem_released, 0);
}
static void __mptcp_splice_receive_queue(struct sock *sk)
@@ -2380,7 +2382,7 @@ static int __mptcp_init_sock(struct sock *sk)
msk->out_of_order_queue = RB_ROOT;
msk->first_pending = NULL;
msk->wmem_reserved = 0;
- msk->rmem_released = 0;
+ WRITE_ONCE(msk->rmem_released, 0);
msk->tx_pending_data = 0;
msk->first = NULL;
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 426ed80fe72f..0f0c026c5f8b 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -296,9 +296,17 @@ static inline struct mptcp_sock *mptcp_sk(const struct sock *sk)
return (struct mptcp_sock *)sk;
}
+/* the msk socket don't use the backlog, also account for the bulk
+ * free memory
+ */
+static inline int __mptcp_rmem(const struct sock *sk)
+{
+ return atomic_read(&sk->sk_rmem_alloc) - READ_ONCE(mptcp_sk(sk)->rmem_released);
+}
+
static inline int __mptcp_space(const struct sock *sk)
{
- return tcp_space(sk) + READ_ONCE(mptcp_sk(sk)->rmem_released);
+ return tcp_win_from_space(sk, READ_ONCE(sk->sk_rcvbuf) - __mptcp_rmem(sk));
}
static inline struct mptcp_data_frag *mptcp_send_head(const struct sock *sk)
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* Re: [PATCH net 0/6] mptcp: Connection and accounting fixes
2021-07-10 0:20 [PATCH net 0/6] mptcp: Connection and accounting fixes Mat Martineau
` (5 preceding siblings ...)
2021-07-10 0:20 ` [PATCH net 6/6] mptcp: properly account bulk freed memory Mat Martineau
@ 2021-07-10 1:50 ` patchwork-bot+netdevbpf
6 siblings, 0 replies; 8+ messages in thread
From: patchwork-bot+netdevbpf @ 2021-07-10 1:50 UTC (permalink / raw)
To: Mat Martineau
Cc: netdev, davem, kuba, matthieu.baerts, pabeni, fw, geliangtang, mptcp
Hello:
This series was applied to netdev/net.git (refs/heads/master):
On Fri, 9 Jul 2021 17:20:45 -0700 you wrote:
> Here are some miscellaneous fixes for MPTCP:
>
> Patch 1 modifies an MPTCP hash so it doesn't depend on one of skb->dev
> and skb->sk being non-NULL.
>
> Patch 2 removes an extra destructor call when rejecting a join due to
> port mismatch.
>
> [...]
Here is the summary with links:
- [net,1/6] mptcp: fix warning in __skb_flow_dissect() when do syn cookie for subflow join
https://git.kernel.org/netdev/net/c/0c71929b5893
- [net,2/6] mptcp: remove redundant req destruct in subflow_check_req()
https://git.kernel.org/netdev/net/c/030d37bd1cd2
- [net,3/6] mptcp: fix syncookie process if mptcp can not_accept new subflow
https://git.kernel.org/netdev/net/c/8547ea5f52dd
- [net,4/6] mptcp: avoid processing packet if a subflow reset
https://git.kernel.org/netdev/net/c/6787b7e350d3
- [net,5/6] selftests: mptcp: fix case multiple subflows limited by server
https://git.kernel.org/netdev/net/c/a7da441621c7
- [net,6/6] mptcp: properly account bulk freed memory
https://git.kernel.org/netdev/net/c/ce599c516386
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2021-07-10 1:50 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-07-10 0:20 [PATCH net 0/6] mptcp: Connection and accounting fixes Mat Martineau
2021-07-10 0:20 ` [PATCH net 1/6] mptcp: fix warning in __skb_flow_dissect() when do syn cookie for subflow join Mat Martineau
2021-07-10 0:20 ` [PATCH net 2/6] mptcp: remove redundant req destruct in subflow_check_req() Mat Martineau
2021-07-10 0:20 ` [PATCH net 3/6] mptcp: fix syncookie process if mptcp can not_accept new subflow Mat Martineau
2021-07-10 0:20 ` [PATCH net 4/6] mptcp: avoid processing packet if a subflow reset Mat Martineau
2021-07-10 0:20 ` [PATCH net 5/6] selftests: mptcp: fix case multiple subflows limited by server Mat Martineau
2021-07-10 0:20 ` [PATCH net 6/6] mptcp: properly account bulk freed memory Mat Martineau
2021-07-10 1:50 ` [PATCH net 0/6] mptcp: Connection and accounting fixes patchwork-bot+netdevbpf
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).