conf->token = talloc_strdup(conf, token); - conf->nr = nat->num_bsc; + conf->nr = number;
I think you could completely remove the num_bsc variable? It looks like its sole use was to determine the next available BSC number without iterating the list.
conf->nat = nat; conf->max_endpoints = 32; conf->paging_group = PAGIN_GROUP_UNASSIGNED; @@ -205,6 +206,10 @@ void bsc_config_free(struct bsc_config *cfg) llist_del(&cfg->entry); rate_ctr_group_free(cfg->stats.ctrg); talloc_free(cfg); + cfg->nat->num_bsc--; + if (cfg->nat->num_bsc < 0) + LOGP(DNAT, LOGL_ERROR, "Internal error while deallocating BSC " + "config: negative BSC index!\n"); }
I don't understand why you would add this check for negative BSC index. The llist_del() should ensure that we don't double free a BSC, right? Also nice would be to add a test case that has a non-null BSC number, to show that having gaps in the numbering doesn't have side effects. ~Neels