/* Accouting handling for netfilter. */ /* * (C) 2008 Krzysztof Piotr Oledzki * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. */ #include #include #include #include #include #if defined(CONFIG_BCM_KF_BLOG) && defined(CONFIG_BLOG) #include #endif #include #include #include #if defined(CONFIG_BCM_KF_DPI) && defined(CONFIG_BCM_DPI_MODULE) static bool nf_ct_acct __read_mostly = 1; #else static bool nf_ct_acct __read_mostly; #endif module_param_named(acct, nf_ct_acct, bool, 0644); MODULE_PARM_DESC(acct, "Enable connection tracking flow accounting."); #ifdef CONFIG_SYSCTL static struct ctl_table acct_sysctl_table[] = { { .procname = "nf_conntrack_acct", .data = &init_net.ct.sysctl_acct, .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec, }, {} }; #endif /* CONFIG_SYSCTL */ unsigned int seq_print_acct(struct seq_file *s, const struct nf_conn *ct, int dir) { struct nf_conn_acct *acct; struct nf_conn_counter *counter; acct = nf_conn_acct_find(ct); if (!acct) return 0; counter = acct->counter; #if defined(CONFIG_BCM_KF_BLOG) && defined(CONFIG_BLOG) { unsigned long long pkts; unsigned long long bytes; FlwStIf_t fast_stats; pkts = (unsigned long long)atomic64_read(&counter[dir].packets); bytes = (unsigned long long)atomic64_read(&counter[dir].bytes); fast_stats.rx_packets = 0; fast_stats.rx_bytes = 0; if (ct->blog_key[dir] != BLOG_KEY_FC_INVALID) { flwStIf_request(FLWSTIF_REQ_GET, &fast_stats, ct->blog_key[dir], 0, 0, 0); counter[dir].ts = fast_stats.pollTS_ms; } seq_printf(s, "packets=%llu bytes=%llu ", pkts+counter[dir].cum_fast_pkts+fast_stats.rx_packets, bytes+counter[dir].cum_fast_bytes+fast_stats.rx_bytes); } #else seq_printf(s, "packets=%llu bytes=%llu ", (unsigned long long)atomic64_read(&counter[dir].packets), (unsigned long long)atomic64_read(&counter[dir].bytes)); #endif return 0; }; EXPORT_SYMBOL_GPL(seq_print_acct); #if defined(CONFIG_BCM_KF_DPI) && defined(CONFIG_BCM_DPI_MODULE) static void conntrack_dpi_refresh_conn_stats(struct nf_conn *ct, int dir, struct dpi_ct_stats *stats) { struct nf_conn_acct *acct; struct nf_conn_counter *counter; #if defined(CONFIG_BCM_KF_BLOG) && defined(CONFIG_BLOG) FlwStIf_t fast_stats; #endif if (!ct->dpi.appinst) return; acct = nf_conn_acct_find(ct); if (!acct) return; counter = acct->counter; /* refresh current connection stats */ stats->pkts = atomic64_read(&counter[dir].packets); stats->bytes = atomic64_read(&counter[dir].bytes); #if defined(CONFIG_BCM_KF_BLOG) && defined(CONFIG_BLOG) fast_stats.rx_packets = 0; fast_stats.rx_bytes = 0; if (ct->blog_key[dir] != BLOG_KEY_FC_INVALID) { flwStIf_request(FLWSTIF_REQ_GET, &fast_stats, ct->blog_key[dir], 0, 0, 0); counter[dir].ts = fast_stats.pollTS_ms; } /* update stats from accelerators */ stats->pkts +=counter[dir].cum_fast_pkts + fast_stats.rx_packets; stats->bytes += counter[dir].cum_fast_bytes + fast_stats.rx_bytes; stats->ts = counter[dir].ts; #endif } void conntrack_dpi_evict_conn(struct nf_conn *ct, int dir) { struct dpi_ct_stats *appinst_stats; struct dpi_ct_stats *dev_stats; struct nf_conn_acct *acct; struct nf_conn_counter *counter; uint64_t packets, bytes; if (!dpi_ops) return; acct = nf_conn_acct_find(ct); if (!acct) return; counter = acct->counter; packets = atomic64_read(&counter[dir].packets); bytes = atomic64_read(&counter[dir].bytes); pr_debug("app<%p> dev<%p> appinst<%p> dir<%s> counter<%llu, %llu>", ct->dpi.app, ct->dpi.dev, ct->dpi.appinst, dir == IP_CT_DIR_ORIGINAL ? "orig" : "reply", packets, bytes); #if defined(CONFIG_BCM_KF_BLOG) && defined(CONFIG_BLOG) pr_debug(" blog_counter<%lu, %llu>\n", counter[dir].cum_fast_pkts, counter[dir].cum_fast_bytes); packets += counter[dir].cum_fast_pkts; bytes += counter[dir].cum_fast_bytes; #endif /* update cumulative appinst stats */ if (ct->dpi.appinst) { appinst_stats = dpi_ops->appinst_stats(ct, dir); appinst_stats->pkts += packets; appinst_stats->bytes += bytes; pr_debug(" appinst_stats<%llu, %llu>", appinst_stats->pkts, appinst_stats->bytes); } /* update cumulative device stats */ if (ct->dpi.dev) { dev_stats = dpi_ops->dev_stats(ct, dir); dev_stats->pkts += packets; dev_stats->bytes += bytes; pr_debug(" dev_stats<%llu, %llu>", dev_stats->pkts, dev_stats->bytes); } pr_debug("\n"); } int conntrack_dpi_seq_print_stats(struct seq_file *s, struct nf_conn *ct, int dir) { struct dpi_ct_stats stats = {0}; /* get updated stats for the connection */ conntrack_dpi_refresh_conn_stats(ct, dir, &stats); return seq_printf(s, " %llu %llu %lu", stats.pkts, stats.bytes, stats.ts); } #endif /* defined(CONFIG_BCM_KF_DPI) && defined(CONFIG_BCM_DPI_MODULE) */ static struct nf_ct_ext_type acct_extend __read_mostly = { .len = sizeof(struct nf_conn_acct), .align = __alignof__(struct nf_conn_acct), .id = NF_CT_EXT_ACCT, }; #ifdef CONFIG_SYSCTL static int nf_conntrack_acct_init_sysctl(struct net *net) { struct ctl_table *table; table = kmemdup(acct_sysctl_table, sizeof(acct_sysctl_table), GFP_KERNEL); if (!table) goto out; table[0].data = &net->ct.sysctl_acct; /* Don't export sysctls to unprivileged users */ if (net->user_ns != &init_user_ns) table[0].procname = NULL; net->ct.acct_sysctl_header = register_net_sysctl(net, "net/netfilter", table); if (!net->ct.acct_sysctl_header) { printk(KERN_ERR "nf_conntrack_acct: can't register to sysctl.\n"); goto out_register; } return 0; out_register: kfree(table); out: return -ENOMEM; } static void nf_conntrack_acct_fini_sysctl(struct net *net) { struct ctl_table *table; table = net->ct.acct_sysctl_header->ctl_table_arg; unregister_net_sysctl_table(net->ct.acct_sysctl_header); kfree(table); } #else static int nf_conntrack_acct_init_sysctl(struct net *net) { return 0; } static void nf_conntrack_acct_fini_sysctl(struct net *net) { } #endif #if defined(CONFIG_BCM_KF_BLOG) && defined(CONFIG_BLOG) /* *--------------------------------------------------------------------------- * Function Name: flwStPushFunc *--------------------------------------------------------------------------- */ int flwStPushFunc( void *ctk1, void *ctk2, uint32_t dir1, uint32_t dir2, FlwStIf_t *flwSt_p ) { struct nf_conn_acct *acct; struct nf_conn_counter *counter; if (flwSt_p == NULL) return -1; if (ctk1 != NULL) { acct = nf_conn_acct_find((struct nf_conn *)ctk1); if (acct) { counter = acct->counter; counter[dir1].cum_fast_pkts += flwSt_p->rx_packets; counter[dir1].cum_fast_bytes += flwSt_p->rx_bytes; counter[dir1].ts = flwSt_p->pollTS_ms; } } if (ctk2 != NULL) { acct = nf_conn_acct_find((struct nf_conn *)ctk2); if (acct) { counter = acct->counter; counter[dir2].cum_fast_pkts += flwSt_p->rx_packets; counter[dir2].cum_fast_bytes += flwSt_p->rx_bytes; counter[dir2].ts = flwSt_p->pollTS_ms; } } return 0; } #endif int nf_conntrack_acct_pernet_init(struct net *net) { net->ct.sysctl_acct = nf_ct_acct; return nf_conntrack_acct_init_sysctl(net); } void nf_conntrack_acct_pernet_fini(struct net *net) { nf_conntrack_acct_fini_sysctl(net); } int nf_conntrack_acct_init(void) { int ret = nf_ct_extend_register(&acct_extend); if (ret < 0) pr_err("nf_conntrack_acct: Unable to register extension\n"); #if defined(CONFIG_BCM_KF_BLOG) && defined(CONFIG_BLOG) flwStIf_bind(NULL, flwStPushFunc); #endif return ret; } void nf_conntrack_acct_fini(void) { nf_ct_extend_unregister(&acct_extend); }