diff options
Diffstat (limited to 'net/core')
-rw-r--r-- | net/core/dev.c | 82 | ||||
-rw-r--r-- | net/core/neighbour.c | 14 | ||||
-rw-r--r-- | net/core/net_namespace.c | 88 | ||||
-rw-r--r-- | net/core/skbuff.c | 99 | ||||
-rw-r--r-- | net/core/sock.c | 3 |
5 files changed, 186 insertions, 100 deletions
diff --git a/net/core/dev.c b/net/core/dev.c index 5f736f1ceeae..72b0d26fd46d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1088,6 +1088,11 @@ int dev_open(struct net_device *dev) dev->flags |= IFF_UP; /* + * Enable NET_DMA + */ + net_dmaengine_get(); + + /* * Initialize multicasting status */ dev_set_rx_mode(dev); @@ -1164,6 +1169,11 @@ int dev_close(struct net_device *dev) */ call_netdevice_notifiers(NETDEV_DOWN, dev); + /* + * Shutdown NET_DMA + */ + net_dmaengine_put(); + return 0; } @@ -1524,7 +1534,19 @@ struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features) skb->mac_len = skb->network_header - skb->mac_header; __skb_pull(skb, skb->mac_len); - if (WARN_ON(skb->ip_summed != CHECKSUM_PARTIAL)) { + if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL)) { + struct net_device *dev = skb->dev; + struct ethtool_drvinfo info = {}; + + if (dev && dev->ethtool_ops && dev->ethtool_ops->get_drvinfo) + dev->ethtool_ops->get_drvinfo(dev, &info); + + WARN(1, "%s: caps=(0x%lx, 0x%lx) len=%d data_len=%d " + "ip_summed=%d", + info.driver, dev ? dev->features : 0L, + skb->sk ? skb->sk->sk_route_caps : 0L, + skb->len, skb->data_len, skb->ip_summed); + if (skb_header_cloned(skb) && (err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC))) return ERR_PTR(err); @@ -2382,6 +2404,9 @@ int dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb) if (!(skb->dev->features & NETIF_F_GRO)) goto normal; + if (skb_is_gso(skb) || skb_shinfo(skb)->frag_list) + goto normal; + rcu_read_lock(); list_for_each_entry_rcu(ptype, head, list) { struct sk_buff *p; @@ -2463,6 +2488,9 @@ static int __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { + if (netpoll_receive_skb(skb)) + return NET_RX_DROP; + switch (__napi_gro_receive(napi, skb)) { case -1: return netif_receive_skb(skb); @@ -2478,12 +2506,6 @@ EXPORT_SYMBOL(napi_gro_receive); void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb) { - skb_shinfo(skb)->nr_frags = 0; - - skb->len -= skb->data_len; - skb->truesize -= skb->data_len; - skb->data_len = 0; - __skb_pull(skb, skb_headlen(skb)); skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb)); @@ -2517,6 +2539,7 @@ struct sk_buff *napi_fraginfo_skb(struct napi_struct *napi, if (!pskb_may_pull(skb, ETH_HLEN)) { napi_reuse_skb(napi, skb); + skb = NULL; goto out; } @@ -2538,6 +2561,9 @@ int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info) if (!skb) goto out; + if (netpoll_receive_skb(skb)) + goto out; + err = NET_RX_SUCCESS; switch (__napi_gro_receive(napi, skb)) { @@ -4424,6 +4450,45 @@ err_uninit: } /** + * init_dummy_netdev - init a dummy network device for NAPI + * @dev: device to init + * + * This takes a network device structure and initialize the minimum + * amount of fields so it can be used to schedule NAPI polls without + * registering a full blown interface. This is to be used by drivers + * that need to tie several hardware interfaces to a single NAPI + * poll scheduler due to HW limitations. + */ +int init_dummy_netdev(struct net_device *dev) +{ + /* Clear everything. Note we don't initialize spinlocks + * are they aren't supposed to be taken by any of the + * NAPI code and this dummy netdev is supposed to be + * only ever used for NAPI polls + */ + memset(dev, 0, sizeof(struct net_device)); + + /* make sure we BUG if trying to hit standard + * register/unregister code path + */ + dev->reg_state = NETREG_DUMMY; + + /* initialize the ref count */ + atomic_set(&dev->refcnt, 1); + + /* NAPI wants this */ + INIT_LIST_HEAD(&dev->napi_list); + + /* a dummy interface is started by default */ + set_bit(__LINK_STATE_PRESENT, &dev->state); + set_bit(__LINK_STATE_START, &dev->state); + + return 0; +} +EXPORT_SYMBOL_GPL(init_dummy_netdev); + + +/** * register_netdev - register a network device * @dev: device to register * @@ -5151,9 +5216,6 @@ static int __init net_dev_init(void) hotcpu_notifier(dev_cpu_callback, 0); dst_init(); dev_mcast_init(); - #ifdef CONFIG_NET_DMA - dmaengine_get(); - #endif rc = 0; out: return rc; diff --git a/net/core/neighbour.c b/net/core/neighbour.c index f66c58df8953..278a142d1047 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1994,8 +1994,8 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) if (!net_eq(neigh_parms_net(p), net)) continue; - if (nidx++ < neigh_skip) - continue; + if (nidx < neigh_skip) + goto next; if (neightbl_fill_param_info(skb, tbl, p, NETLINK_CB(cb->skb).pid, @@ -2003,6 +2003,8 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) RTM_NEWNEIGHTBL, NLM_F_MULTI) <= 0) goto out; + next: + nidx++; } neigh_skip = 0; @@ -2082,12 +2084,10 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, if (h > s_h) s_idx = 0; for (n = tbl->hash_buckets[h], idx = 0; n; n = n->next) { - int lidx; if (dev_net(n->dev) != net) continue; - lidx = idx++; - if (lidx < s_idx) - continue; + if (idx < s_idx) + goto next; if (neigh_fill_info(skb, n, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, RTM_NEWNEIGH, @@ -2096,6 +2096,8 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, rc = -1; goto out; } + next: + idx++; } } read_unlock_bh(&tbl->lock); diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 55cffad2f328..2adb1a7d361f 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -32,24 +32,14 @@ static __net_init int setup_net(struct net *net) { /* Must be called with net_mutex held */ struct pernet_operations *ops; - int error; - struct net_generic *ng; + int error = 0; atomic_set(&net->count, 1); + #ifdef NETNS_REFCNT_DEBUG atomic_set(&net->use_count, 0); #endif - error = -ENOMEM; - ng = kzalloc(sizeof(struct net_generic) + - INITIAL_NET_GEN_PTRS * sizeof(void *), GFP_KERNEL); - if (ng == NULL) - goto out; - - ng->len = INITIAL_NET_GEN_PTRS; - rcu_assign_pointer(net->gen, ng); - - error = 0; list_for_each_entry(ops, &pernet_list, list) { if (ops->init) { error = ops->init(net); @@ -70,24 +60,50 @@ out_undo: } rcu_barrier(); - kfree(ng); goto out; } +static struct net_generic *net_alloc_generic(void) +{ + struct net_generic *ng; + size_t generic_size = sizeof(struct net_generic) + + INITIAL_NET_GEN_PTRS * sizeof(void *); + + ng = kzalloc(generic_size, GFP_KERNEL); + if (ng) + ng->len = INITIAL_NET_GEN_PTRS; + + return ng; +} + #ifdef CONFIG_NET_NS static struct kmem_cache *net_cachep; static struct workqueue_struct *netns_wq; static struct net *net_alloc(void) { - return kmem_cache_zalloc(net_cachep, GFP_KERNEL); + struct net *net = NULL; + struct net_generic *ng; + + ng = net_alloc_generic(); + if (!ng) + goto out; + + net = kmem_cache_zalloc(net_cachep, GFP_KERNEL); + if (!net) + goto out_free; + + rcu_assign_pointer(net->gen, ng); +out: + return net; + +out_free: + kfree(ng); + goto out; } static void net_free(struct net *net) { - if (!net) - return; - #ifdef NETNS_REFCNT_DEBUG if (unlikely(atomic_read(&net->use_count) != 0)) { printk(KERN_EMERG "network namespace not free! Usage: %d\n", @@ -112,27 +128,28 @@ struct net *copy_net_ns(unsigned long flags, struct net *old_net) err = -ENOMEM; new_net = net_alloc(); if (!new_net) - goto out; + goto out_err; mutex_lock(&net_mutex); err = setup_net(new_net); - if (err) - goto out_unlock; - - rtnl_lock(); - list_add_tail(&new_net->list, &net_namespace_list); - rtnl_unlock(); - - -out_unlock: + if (!err) { + rtnl_lock(); + list_add_tail(&new_net->list, &net_namespace_list); + rtnl_unlock(); + } mutex_unlock(&net_mutex); + + if (err) + goto out_free; out: put_net(old_net); - if (err) { - net_free(new_net); - new_net = ERR_PTR(err); - } return new_net; + +out_free: + net_free(new_net); +out_err: + new_net = ERR_PTR(err); + goto out; } static void cleanup_net(struct work_struct *work) @@ -188,6 +205,7 @@ struct net *copy_net_ns(unsigned long flags, struct net *old_net) static int __init net_ns_init(void) { + struct net_generic *ng; int err; printk(KERN_INFO "net_namespace: %zd bytes\n", sizeof(struct net)); @@ -202,6 +220,12 @@ static int __init net_ns_init(void) panic("Could not create netns workq"); #endif + ng = net_alloc_generic(); + if (!ng) + panic("Could not allocate generic netns"); + + rcu_assign_pointer(init_net.gen, ng); + mutex_lock(&net_mutex); err = setup_net(&init_net); @@ -341,8 +365,8 @@ again: rv = register_pernet_operations(first_device, ops); if (rv < 0) ida_remove(&net_generic_ids, *id); - mutex_unlock(&net_mutex); out: + mutex_unlock(&net_mutex); return rv; } EXPORT_SYMBOL_GPL(register_pernet_gen_subsys); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 5110b359c758..c6a6b166f8d6 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -73,17 +73,13 @@ static struct kmem_cache *skbuff_fclone_cache __read_mostly; static void sock_pipe_buf_release(struct pipe_inode_info *pipe, struct pipe_buffer *buf) { - struct sk_buff *skb = (struct sk_buff *) buf->private; - - kfree_skb(skb); + put_page(buf->page); } static void sock_pipe_buf_get(struct pipe_inode_info *pipe, struct pipe_buffer *buf) { - struct sk_buff *skb = (struct sk_buff *) buf->private; - - skb_get(skb); + get_page(buf->page); } static int sock_pipe_buf_steal(struct pipe_inode_info *pipe, @@ -147,14 +143,6 @@ void skb_under_panic(struct sk_buff *skb, int sz, void *here) BUG(); } -void skb_truesize_bug(struct sk_buff *skb) -{ - WARN(net_ratelimit(), KERN_ERR "SKB BUG: Invalid truesize (%u) " - "len=%u, sizeof(sk_buff)=%Zd\n", - skb->truesize, skb->len, sizeof(struct sk_buff)); -} -EXPORT_SYMBOL(skb_truesize_bug); - /* Allocate a new skbuff. We do this ourselves so we can fill in a few * 'private' fields and also do memory statistics to find all the * [BEEP] leaks. @@ -1334,9 +1322,19 @@ fault: */ static void sock_spd_release(struct splice_pipe_desc *spd, unsigned int i) { - struct sk_buff *skb = (struct sk_buff *) spd->partial[i].private; + put_page(spd->pages[i]); +} - kfree_skb(skb); +static inline struct page *linear_to_page(struct page *page, unsigned int len, + unsigned int offset) +{ + struct page *p = alloc_pages(GFP_KERNEL, 0); + + if (!p) + return NULL; + memcpy(page_address(p) + offset, page_address(page) + offset, len); + + return p; } /* @@ -1344,16 +1342,23 @@ static void sock_spd_release(struct splice_pipe_desc *spd, unsigned int i) */ static inline int spd_fill_page(struct splice_pipe_desc *spd, struct page *page, unsigned int len, unsigned int offset, - struct sk_buff *skb) + struct sk_buff *skb, int linear) { if (unlikely(spd->nr_pages == PIPE_BUFFERS)) return 1; + if (linear) { + page = linear_to_page(page, len, offset); + if (!page) + return 1; + } else + get_page(page); + spd->pages[spd->nr_pages] = page; spd->partial[spd->nr_pages].len = len; spd->partial[spd->nr_pages].offset = offset; - spd->partial[spd->nr_pages].private = (unsigned long) skb_get(skb); spd->nr_pages++; + return 0; } @@ -1369,7 +1374,7 @@ static inline void __segment_seek(struct page **page, unsigned int *poff, static inline int __splice_segment(struct page *page, unsigned int poff, unsigned int plen, unsigned int *off, unsigned int *len, struct sk_buff *skb, - struct splice_pipe_desc *spd) + struct splice_pipe_desc *spd, int linear) { if (!*len) return 1; @@ -1392,7 +1397,7 @@ static inline int __splice_segment(struct page *page, unsigned int poff, /* the linear region may spread across several pages */ flen = min_t(unsigned int, flen, PAGE_SIZE - poff); - if (spd_fill_page(spd, page, flen, poff, skb)) + if (spd_fill_page(spd, page, flen, poff, skb, linear)) return 1; __segment_seek(&page, &poff, &plen, flen); @@ -1419,7 +1424,7 @@ static int __skb_splice_bits(struct sk_buff *skb, unsigned int *offset, if (__splice_segment(virt_to_page(skb->data), (unsigned long) skb->data & (PAGE_SIZE - 1), skb_headlen(skb), - offset, len, skb, spd)) + offset, len, skb, spd, 1)) return 1; /* @@ -1429,7 +1434,7 @@ static int __skb_splice_bits(struct sk_buff *skb, unsigned int *offset, const skb_frag_t *f = &skb_shinfo(skb)->frags[seg]; if (__splice_segment(f->page, f->page_offset, f->size, - offset, len, skb, spd)) + offset, len, skb, spd, 0)) return 1; } @@ -1442,7 +1447,7 @@ static int __skb_splice_bits(struct sk_buff *skb, unsigned int *offset, * the frag list, if such a thing exists. We'd probably need to recurse to * handle that cleanly. */ -int skb_splice_bits(struct sk_buff *__skb, unsigned int offset, +int skb_splice_bits(struct sk_buff *skb, unsigned int offset, struct pipe_inode_info *pipe, unsigned int tlen, unsigned int flags) { @@ -1455,16 +1460,6 @@ int skb_splice_bits(struct sk_buff *__skb, unsigned int offset, .ops = &sock_pipe_buf_ops, .spd_release = sock_spd_release, }; - struct sk_buff *skb; - - /* - * I'd love to avoid the clone here, but tcp_read_sock() - * ignores reference counts and unconditonally kills the sk_buff - * on return from the actor. - */ - skb = skb_clone(__skb, GFP_KERNEL); - if (unlikely(!skb)) - return -ENOMEM; /* * __skb_splice_bits() only fails if the output has no room left, @@ -1488,15 +1483,9 @@ int skb_splice_bits(struct sk_buff *__skb, unsigned int offset, } done: - /* - * drop our reference to the clone, the pipe consumption will - * drop the rest. - */ - kfree_skb(skb); - if (spd.nr_pages) { + struct sock *sk = skb->sk; int ret; - struct sock *sk = __skb->sk; /* * Drop the socket lock, otherwise we have reverse @@ -2215,10 +2204,10 @@ unsigned int skb_seq_read(unsigned int consumed, const u8 **data, return 0; next_skb: - block_limit = skb_headlen(st->cur_skb); + block_limit = skb_headlen(st->cur_skb) + st->stepped_offset; if (abs_offset < block_limit) { - *data = st->cur_skb->data + abs_offset; + *data = st->cur_skb->data + (abs_offset - st->stepped_offset); return block_limit - abs_offset; } @@ -2253,13 +2242,14 @@ next_skb: st->frag_data = NULL; } - if (st->cur_skb->next) { - st->cur_skb = st->cur_skb->next; + if (st->root_skb == st->cur_skb && + skb_shinfo(st->root_skb)->frag_list) { + st->cur_skb = skb_shinfo(st->root_skb)->frag_list; st->frag_idx = 0; goto next_skb; - } else if (st->root_skb == st->cur_skb && - skb_shinfo(st->root_skb)->frag_list) { - st->cur_skb = skb_shinfo(st->root_skb)->frag_list; + } else if (st->cur_skb->next) { + st->cur_skb = st->cur_skb->next; + st->frag_idx = 0; goto next_skb; } @@ -2588,8 +2578,9 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb) struct sk_buff *nskb; unsigned int headroom; unsigned int hlen = p->data - skb_mac_header(p); + unsigned int len = skb->len; - if (hlen + p->len + skb->len >= 65536) + if (hlen + p->len + len >= 65536) return -E2BIG; if (skb_shinfo(p)->frag_list) @@ -2602,6 +2593,12 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb) skb_shinfo(skb)->nr_frags * sizeof(skb_frag_t)); skb_shinfo(p)->nr_frags += skb_shinfo(skb)->nr_frags; + skb_shinfo(skb)->nr_frags = 0; + + skb->truesize -= skb->data_len; + skb->len -= skb->data_len; + skb->data_len = 0; + NAPI_GRO_CB(skb)->free = 1; goto done; } @@ -2645,9 +2642,9 @@ merge: done: NAPI_GRO_CB(p)->count++; - p->data_len += skb->len; - p->truesize += skb->len; - p->len += skb->len; + p->data_len += len; + p->truesize += len; + p->len += len; NAPI_GRO_CB(skb)->same_flow = 1; return 0; diff --git a/net/core/sock.c b/net/core/sock.c index f3a0d08cbb48..5f97caa158e8 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -696,6 +696,8 @@ int sock_getsockopt(struct socket *sock, int level, int optname, if (len < 0) return -EINVAL; + memset(&v, 0, sizeof(v)); + switch(optname) { case SO_DEBUG: v.val = sock_flag(sk, SOCK_DBG); @@ -1135,7 +1137,6 @@ void sock_rfree(struct sk_buff *skb) { struct sock *sk = skb->sk; - skb_truesize_check(skb); atomic_sub(skb->truesize, &sk->sk_rmem_alloc); sk_mem_uncharge(skb->sk, skb->truesize); } |