diff options
author | Davide Caratti <dcaratti@redhat.com> | 2020-10-06 18:26:17 +0200 |
---|---|---|
committer | Jakub Kicinski <kuba@kernel.org> | 2020-10-09 08:25:48 -0700 |
commit | 37198e93ced70733f0b993dff28b7c33857e254f (patch) | |
tree | 5efe9e0415abf424ed26afd3563052862864ebe4 /net/mptcp/subflow.c | |
parent | 0da1ccbbefb662915228bc17e1c7d4ad28b3ddab (diff) | |
download | lwn-37198e93ced70733f0b993dff28b7c33857e254f.tar.gz lwn-37198e93ced70733f0b993dff28b7c33857e254f.zip |
net: mptcp: make DACK4/DACK8 usage consistent among all subflows
using packetdrill it's possible to observe the same MPTCP DSN being acked
by different subflows with DACK4 and DACK8. This is in contrast with what
specified in RFC8684 §3.3.2: if an MPTCP endpoint transmits a 64-bit wide
DSN, it MUST be acknowledged with a 64-bit wide DACK. Fix 'use_64bit_ack'
variable to make it a property of MPTCP sockets, not TCP subflows.
Fixes: a0c1d0eafd1e ("mptcp: Use 32-bit DATA_ACK when possible")
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Diffstat (limited to 'net/mptcp/subflow.c')
-rw-r--r-- | net/mptcp/subflow.c | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 6f035af1c9d2..91bef7bfffa6 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -769,12 +769,11 @@ static enum mapping_status get_mapping_status(struct sock *ssk, if (!mpext->dsn64) { map_seq = expand_seq(subflow->map_seq, subflow->map_data_len, mpext->data_seq); - subflow->use_64bit_ack = 0; pr_debug("expanded seq=%llu", subflow->map_seq); } else { map_seq = mpext->data_seq; - subflow->use_64bit_ack = 1; } + WRITE_ONCE(mptcp_sk(subflow->conn)->use_64bit_ack, !!mpext->dsn64); if (subflow->map_valid) { /* Allow replacing only with an identical map */ |