| Message ID | 20211122112205.16752-1-fw@strlen.de (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Delegated to: | Matthieu Baerts |
| Headers | show |
| Series | [mptcp-next,1/2] mptcp: getsockopt: add support for IP_TOS | expand |
Hi Florian, On 22/11/2021 12:22, Florian Westphal wrote: > earlier patch added IP_TOS setsockopt support, this allows to get > the value set by earlier setsockopt. Thank you for sharing the patches, good idea to add this! > diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c > index 11cda8629993..5626b49c0270 100644 > --- a/net/mptcp/sockopt.c > +++ b/net/mptcp/sockopt.c > @@ -1066,6 +1066,24 @@ static int mptcp_put_int_option(struct mptcp_sock *msk, char __user *optval, > return 0; > } > > +static int mptcp_put_int8_option(struct mptcp_sock *msk, char __user *optval, > + int __user *optlen, u8 val) > +{ > + int len; > + > + if (get_user(len, optlen)) > + return -EFAULT; > + > + if (len) { What can be the possible values of 'len' here? Can it be negative? I'm asking this question because when I'm looking at similar code in mptcp/sockopt.c, I see different behaviours: - mptcp_getsockopt_info(): no additional check but we suppose 'len' is unsigned - mptcp_put_int_option(): same but we return an error if len < 0 - mptcp_get_subflow_data(): we check if len is not < the expected size In ipv4/ip_sockglue.c, we check if len < 0 I don't know if we should uniform that and always do: if (len < 0 (or expected value)) return (...); But I'm fine with the current version if we would never have a negative value. Cheers, Matt
diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 11cda8629993..5626b49c0270 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -1066,6 +1066,24 @@ static int mptcp_put_int_option(struct mptcp_sock *msk, char __user *optval, return 0; } +static int mptcp_put_int8_option(struct mptcp_sock *msk, char __user *optval, + int __user *optlen, u8 val) +{ + int len; + + if (get_user(len, optlen)) + return -EFAULT; + + if (len) { + if (put_user(1, optlen)) + return -EFAULT; + if (copy_to_user(optval, &val, 1)) + return -EFAULT; + } + + return 0; +} + static int mptcp_getsockopt_sol_tcp(struct mptcp_sock *msk, int optname, char __user *optval, int __user *optlen) { @@ -1082,6 +1100,19 @@ static int mptcp_getsockopt_sol_tcp(struct mptcp_sock *msk, int optname, return -EOPNOTSUPP; } +static int mptcp_getsockopt_v4(struct mptcp_sock *msk, int optname, + char __user *optval, int __user *optlen) +{ + struct sock *sk = (void *)msk; + + switch (optname) { + case IP_TOS: + return mptcp_put_int8_option(msk, optval, optlen, inet_sk(sk)->tos); + } + + return -EOPNOTSUPP; +} + static int mptcp_getsockopt_sol_mptcp(struct mptcp_sock *msk, int optname, char __user *optval, int __user *optlen) { @@ -1117,6 +1148,8 @@ int mptcp_getsockopt(struct sock *sk, int level, int optname, if (ssk) return tcp_getsockopt(ssk, level, optname, optval, option); + if (level == SOL_IP) + return mptcp_getsockopt_v4(msk, optname, optval, option); if (level == SOL_TCP) return mptcp_getsockopt_sol_tcp(msk, optname, optval, option); if (level == SOL_MPTCP)
earlier patch added IP_TOS setsockopt support, this allows to get the value set by earlier setsockopt. Signed-off-by: Florian Westphal <fw@strlen.de> --- net/mptcp/sockopt.c | 33 +++++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+)