On Mon, Aug 19, 2019 at 10:58 PM Marcelo Ricardo Leitner <marcelo.leit...@gmail.com> wrote: > > On Mon, Aug 19, 2019 at 10:02:46PM +0800, Xin Long wrote: > > +static int sctp_setsockopt_asconf_supported(struct sock *sk, > > + char __user *optval, > > + unsigned int optlen) > > +{ > > + struct sctp_assoc_value params; > > + struct sctp_association *asoc; > > + struct sctp_endpoint *ep; > > + int retval = -EINVAL; > > + > > + if (optlen != sizeof(params)) > > + goto out; > > + > > + if (copy_from_user(¶ms, optval, optlen)) { > > + retval = -EFAULT; > > + goto out; > > + } > > + > > + asoc = sctp_id2assoc(sk, params.assoc_id); > > + if (!asoc && params.assoc_id != SCTP_FUTURE_ASSOC && > > + sctp_style(sk, UDP)) > > + goto out; > > + > > + ep = sctp_sk(sk)->ep; > > + ep->asconf_enable = !!params.assoc_value; > > Considering this will be negotiated on handshake, shouldn't it deny > changes to Established asocs? (Same for Auth) ep->asconf_enable is for 'furture' asocs, and furture asoc->peer.asconf_capable will be negotiated according to peer ep->asconf_enable (EXT chunk) and local ep->asconf_enable.
It won't affect the 'current'/Established asocs, and the Established asocs have asoc->peer.asconf_capable, which can't be changed by sockopt. > > > + > > + if (ep->asconf_enable && ep->auth_enable) { > > + sctp_auth_ep_add_chunkid(ep, SCTP_CID_ASCONF); > > + sctp_auth_ep_add_chunkid(ep, SCTP_CID_ASCONF_ACK); > > + } > > + > > + retval = 0; > > + > > +out: > > + return retval; > > +}