Linux内核中流量控制(9)
本文档的Copyleft归yfydz所有，使用GPL发布，可以自由拷贝，转载，转载时请保持文档的完整性，严禁用于任何商业用途。
msn: yfydz_no1@hotmail.com
来源：http://yfydz.cublog.cn
5.9 ingressingress流控方法是针对输入数据进行流控处理的，在net/sched/sch_ingress.c中定义, 使用时要求在内核配置中定义CONFIG_NET_CLS_ACT或CONFIG_NETFILTER，不过从代码实现看起来好象没什么意义。5.9.1 ingress操作结构定义// ingress私有数据结构struct ingress_qdisc_data {// 内部流控 struct Qdisc *q;// 过滤规则 struct tcf_proto *filter_list;};// ingress流控操作结构static struct Qdisc_ops ingress_qdisc_ops = { .next = NULL, .cl_ops = &ingress_class_ops, .id = "ingress", .priv_size = sizeof(struct ingress_qdisc_data), .enqueue = ingress_enqueue, .dequeue = ingress_dequeue, .requeue = ingress_requeue, .drop = ingress_drop, .init = ingress_init, .reset = ingress_reset, .destroy = ingress_destroy,// 无change函数 .change = NULL, .dump = ingress_dump, .owner = THIS_MODULE,}; // ingress类别操作结构static struct Qdisc_class_ops ingress_class_ops = { .graft = ingress_graft, .leaf = ingress_leaf, .get = ingress_get, .put = ingress_put, .change = ingress_change, .delete = NULL, .walk = ingress_walk, .tcf_chain = ingress_find_tcf, .bind_tcf = ingress_bind_filter, .unbind_tcf = ingress_put, .dump = NULL,};5.9.1 ingress的netfilter节点// 没定义NET_CLS_ACT, 而定义了NETFILTER的情况下才有效, 在netfilter节点中进行// 数据包是否接收的判断// 但如果定义了NET_CLS_ACT后, 该段代码无效#ifndef CONFIG_NET_CLS_ACT#ifdef CONFIG_NETFILTER/* after ipt_filter */static struct nf_hook_ops ing_ops = {// 钩子函数 .hook = ing_hook, .owner = THIS_MODULE, .pf = PF_INET,// 挂接在PREROUTING点 .hooknum = NF_IP_PRE_ROUTING,// 优先权NF_IP_PRI_FILTER + 1, 不过在PREROUTING点缺省好象没定义NF_IP_PRI_FILTER// 级别的节点 .priority = NF_IP_PRI_FILTER + 1,};static struct nf_hook_ops ing6_ops = { .hook = ing_hook, .owner = THIS_MODULE, .pf = PF_INET6, .hooknum = NF_IP6_PRE_ROUTING, .priority = NF_IP6_PRI_FILTER + 1,};static unsigned inting_hook(unsigned int hook, struct sk_buff **pskb, const struct net_device *indev, const struct net_device *outdev, int (*okfn)(struct sk_buff *)){ struct Qdisc *q; struct sk_buff *skb = *pskb; struct net_device *dev = skb->dev;// 缺省动作, 接受 int fwres=NF_ACCEPT; DPRINTK("ing_hook: skb %s dev=%s len=%u\n", skb->sk ? "(owned)" : "(unowned)", skb->dev ? (*pskb)->dev->name : "(no dev)", skb->len);/*revisit later: Use a private since lock dev->queue_lock is alsoused on the egress (might slow things for an iota)*/// 如果设备里定义了接收流控 if (dev->qdisc_ingress) { spin_lock(&dev->queue_lock); if ((q = dev->qdisc_ingress) != NULL)// 调用接收流控的入队函数, 返回值是对数据包的判断结果 fwres = q->enqueue(skb, q); spin_unlock(&dev->queue_lock); }// 如果没定义, 就直接返回缺省动作: 通过 return fwres;}#endif#endif5.9.3 初始化static int ingress_init(struct Qdisc *sch,struct rtattr *opt){ struct ingress_qdisc_data *p = PRIV(sch);/* Make sure either netfilter or preferably CLS_ACT is* compiled in */// 输入接收流控必须在内核选项中定义分类动作NET_CLS_ACT或NETFILTER#ifndef CONFIG_NET_CLS_ACT#ifndef CONFIG_NETFILTER// 两者都没定义的话就初始化失败 printk("You MUST compile classifier actions into the kernel\n"); return -EINVAL;#else// 使用NET_CLS_ACT优于使用NETFILTER printk("Ingress scheduler: Classifier actions prefered over netfilter\n");#endif#endif #ifndef CONFIG_NET_CLS_ACT#ifdef CONFIG_NETFILTER// 没定义NET_CLS_ACT, 而定义了NETFILTER的情况// 登记输入流控的netfilter钩子节点函数 if (!nf_registered) {// 登记IPV4钩子节点 if (nf_register_hook(&ing_ops) < 0) { printk("ingress qdisc registration error \n"); return -EINVAL; }// 非0表示已经登记了 nf_registered++;// 登记IPV4钩子节点 if (nf_register_hook(&ing6_ops) < 0) { printk("IPv6 ingress qdisc registration error, " \ "disabling IPv6 support.\n"); } else nf_registered++; }#endif#endif DPRINTK("ingress_init(sch %p,[qdisc %p],opt %p)\n",sch,p,opt);// 初始内部流控初始化为noop, 丢包的 p->q = &noop_qdisc; return 0;} 5.9.5 入队sstatic int ingress_enqueue(struct sk_buff *skb,struct Qdisc *sch){// 接收流控私有数据 struct ingress_qdisc_data *p = PRIV(sch);// 分类结果 struct tcf_result res; int result; D2PRINTK("ingress_enqueue(skb %p,sch %p,[qdisc %p])\n", skb, sch, p);// 对数据进行分类 result = tc_classify(skb, p->filter_list, &res); D2PRINTK("result %d class 0x%04x\n", result, res.classid); /* * Unlike normal "enqueue" functions, ingress_enqueue returns a * firewall FW_* code. */#ifdef CONFIG_NET_CLS_ACT// 在定义了NET_CLS_ACT的情况, 这时不会返回NF_*, 而是返回TC_ACT_* sch->bstats.packets++; sch->bstats.bytes += skb->len; switch (result) { case TC_ACT_SHOT:// 丢包 result = TC_ACT_SHOT; sch->qstats.drops++; break; case TC_ACT_STOLEN: case TC_ACT_QUEUED:// 偷包, 重新处理 result = TC_ACT_STOLEN; break; case TC_ACT_RECLASSIFY: case TC_ACT_OK: case TC_ACT_UNSPEC: default:// 接受, 计算tc_index skb->tc_index = TC_H_MIN(res.classid); result = TC_ACT_OK; break; };/* backward compat */#else// 在没有定义了NET_CLS_ACT的情况, 这时存在netfilter节点, 返回NF_*结果#ifdef CONFIG_NET_CLS_POLICE // 在定义了NET_CLS_POLICE的情况 switch (result) { case TC_POLICE_SHOT:// 丢包 result = NF_DROP; sch->qstats.drops++; break; case TC_POLICE_RECLASSIFY: /* DSCP remarking here ? */ case TC_POLICE_OK: case TC_POLICE_UNSPEC: default:// 其他情况都是ACCEPT sch->bstats.packets++; sch->bstats.bytes += skb->len; result = NF_ACCEPT; break; };#else// 那两个NET_CLS选项都没定义的话直接ACCEPT D2PRINTK("Overriding result to ACCEPT\n"); result = NF_ACCEPT; sch->bstats.packets++; sch->bstats.bytes += skb->len;#endif#endif return result;} 5.9.6 重入队空函数static int ingress_requeue(struct sk_buff *skb,struct Qdisc *sch){/* struct ingress_qdisc_data *p = PRIV(sch); D2PRINTK("ingress_requeue(skb %p,sch %p,[qdisc %p])\n",skb,sch,PRIV(p));*/ return 0;} 5.9.7 出队// 空函数,因为不会有真正的dequeue操作static struct sk_buff *ingress_dequeue(struct Qdisc *sch){/* struct ingress_qdisc_data *p = PRIV(sch); D2PRINTK("ingress_dequeue(sch %p,[qdisc %p])\n",sch,PRIV(p));*/ return NULL;} 5.9.8 丢包// 基本也是空函数static unsigned int ingress_drop(struct Qdisc *sch){#ifdef DEBUG_INGRESS struct ingress_qdisc_data *p = PRIV(sch);#endif DPRINTK("ingress_drop(sch %p,[qdisc %p])\n", sch, p); return 0;} 5.9.9 复位sstatic void ingress_reset(struct Qdisc *sch){ struct ingress_qdisc_data *p = PRIV(sch); DPRINTK("ingress_reset(sch %p,[qdisc %p])\n", sch, p);/*#if 0*//* for future use */// 复位内部流控节点 qdisc_reset(p->q);/*#endif*/} 5.9.10 释放static void ingress_destroy(struct Qdisc *sch){// 私有数据 struct ingress_qdisc_data *p = PRIV(sch); struct tcf_proto *tp; DPRINTK("ingress_destroy(sch %p,[qdisc %p])\n", sch, p);// 释放TC过滤规则 while (p->filter_list) { tp = p->filter_list; p->filter_list = tp->next; tcf_destroy(tp); }#if 0/* for future use */ qdisc_destroy(p->q);#endif}5.9.11 输出参数static int ingress_dump(struct Qdisc *sch, struct sk_buff *skb){ unsigned char *b = skb->tail; struct rtattr *rta; rta = (struct rtattr *) b;// 起始什么数据也没有 RTA_PUT(skb, TCA_OPTIONS, 0, NULL); rta->rta_len = skb->tail - b; return skb->len;rtattr_failure: skb_trim(skb, b - skb->data); return -1;}5.9.13 ingress类别操作// 嫁接, 增加叶子节点// 基本是空函数, 恒返回1static int ingress_graft(struct Qdisc *sch,unsigned long arg, struct Qdisc *new,struct Qdisc **old){#ifdef DEBUG_INGRESS struct ingress_qdisc_data *p = PRIV(sch);#endif DPRINTK("ingress_graft(sch %p,[qdisc %p],new %p,old %p)\n", sch, p, new, old); DPRINTK("\n ingress_graft: You cannot add qdiscs to classes"); return 1;}// 叶子, 空函数static struct Qdisc *ingress_leaf(struct Qdisc *sch, unsigned long arg){ return NULL;}// 计数增加, 使用类别ID计算static unsigned long ingress_get(struct Qdisc *sch,u32 classid){#ifdef DEBUG_INGRESS struct ingress_qdisc_data *p = PRIV(sch);#endif DPRINTK("ingress_get(sch %p,[qdisc %p],classid %x)\n", sch, p, classid); return TC_H_MIN(classid) + 1;}// 绑定TC过滤规则表static unsigned long ingress_bind_filter(struct Qdisc *sch, unsigned long parent, u32 classid){ return ingress_get(sch, classid);}// 计数减少, 空函数static void ingress_put(struct Qdisc *sch, unsigned long cl){}// 空函数static int ingress_change(struct Qdisc *sch, u32 classid, u32 parent, struct rtattr **tca, unsigned long *arg){#ifdef DEBUG_INGRESS struct ingress_qdisc_data *p = PRIV(sch);#endif DPRINTK("ingress_change(sch %p,[qdisc %p],classid %x,parent %x)," "arg 0x%lx\n", sch, p, classid, parent, *arg); DPRINTK("No effect. sch_ingress doesn't maintain classes at the moment"); return 0;} // 遍历, 空函数static void ingress_walk(struct Qdisc *sch,struct qdisc_walker *walker){#ifdef DEBUG_INGRESS struct ingress_qdisc_data *p = PRIV(sch);#endif DPRINTK("ingress_walk(sch %p,[qdisc %p],walker %p)\n", sch, p, walker); DPRINTK("No effect. sch_ingress doesn't maintain classes at the moment");}// 获取TC过滤规则表static struct tcf_proto **ingress_find_tcf(struct Qdisc *sch,unsigned long cl){ struct ingress_qdisc_data *p = PRIV(sch); return &p->filter_list;} ...... 待续 ......