BUF_ASSERT_USED(buf);
- /* XXX crunchy bytestuff code */
+ /* There's probably a much more efficient way of implementing this */
txbuf[j++]=SLIP_END;
for (i=buf->start; i<(buf->start+buf->size); i++) {
switch (*i) {
{
uint32_t i;
- /* XXX really crude unstuff code */
- /* XXX check for buffer overflow */
BUF_ASSERT_USED(st->buff);
for (i=0; i<l; i++) {
if (st->pending_esc) {
string_t addrs;
string_t nets;
string_t s;
- struct netlink_route *r;
- struct ipset *isnets;
+ struct netlink_client *r;
+ struct ipset *allnets;
struct subnet_list *snets;
int i;
uint8_t confirm;
ipaddr_to_string(st->slip.local_address),
ipaddr_to_string(st->slip.nl.secnet_address),st->slip.nl.mtu);
- r=st->slip.nl.routes;
- isnets=ipset_new();
- for (i=0; i<st->slip.nl.n_routes; i++) {
- if (r[i].up) {
- struct ipset *sn,*nis;
- r[i].kup=True;
- sn=ipset_from_subnet(r[i].net);
- nis=ipset_union(isnets,sn);
- ipset_free(sn);
- ipset_free(isnets);
- isnets=nis;
+ allnets=ipset_new();
+ for (r=st->slip.nl.clients; r; r=r->next) {
+ if (r->up) {
+ struct ipset *nan;
+ r->kup=True;
+ nan=ipset_union(allnets,r->networks);
+ ipset_free(allnets);
+ allnets=nan;
}
}
- snets=ipset_to_subnet_list(isnets);
- ipset_free(isnets);
+ snets=ipset_to_subnet_list(allnets);
+ ipset_free(allnets);
nets=safe_malloc(20*snets->entries,"userv_invoke_userv:nets");
*nets=0;
for (i=0; i<snets->entries; i++) {
void slip_module(dict_t *dict)
{
add_closure(dict,"userv-ipif",userv_apply);
-#if 0
- /* TODO */
- add_closure(dict,"pty-slip",ptyslip_apply);
- add_closure(dict,"slipd",slipd_apply);
-#endif /* 0 */
}