From 1716a96101c49186bb0b8491922fd3e69030235f Mon Sep 17 00:00:00 2001 From: Gao feng Date: Fri, 6 Apr 2012 00:13:10 +0000 Subject: [PATCH 01/11] ipv6: fix problem with expired dst cache If the ipv6 dst cache which copy from the dst generated by ICMPV6 RA packet. this dst cache will not check expire because it has no RTF_EXPIRES flag. So this dst cache will always be used until the dst gc run. Change the struct dst_entry,add a union contains new pointer from and expires. When rt6_info.rt6i_flags has no RTF_EXPIRES flag,the dst.expires has no use. we can use this field to point to where the dst cache copy from. The dst.from is only used in IPV6. rt6_check_expired check if rt6_info.dst.from is expired. ip6_rt_copy only set dst.from when the ort has flag RTF_ADDRCONF and RTF_DEFAULT.then hold the ort. ip6_dst_destroy release the ort. Add some functions to operate the RTF_EXPIRES flag and expires(from) together. and change the code to use these new adding functions. Changes from v5: modify ip6_route_add and ndisc_router_discovery to use new adding functions. Only set dst.from when the ort has flag RTF_ADDRCONF and RTF_DEFAULT.then hold the ort. Signed-off-by: Gao feng Signed-off-by: David S. Miller --- include/net/dst.h | 6 +++- include/net/ip6_fib.h | 42 +++++++++++++++++++++++++ net/ipv6/addrconf.c | 9 ++---- net/ipv6/ip6_fib.c | 9 +++--- net/ipv6/ndisc.c | 3 +- net/ipv6/route.c | 71 +++++++++++++++++++++++++++---------------- 6 files changed, 99 insertions(+), 41 deletions(-) diff --git a/include/net/dst.h b/include/net/dst.h index 59c5d18cc385..ff4da42fcfc6 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -36,7 +36,11 @@ struct dst_entry { struct net_device *dev; struct dst_ops *ops; unsigned long _metrics; - unsigned long expires; + union { + unsigned long expires; + /* point to where the dst_entry copied from */ + struct dst_entry *from; + }; struct dst_entry *path; struct neighbour __rcu *_neighbour; #ifdef CONFIG_XFRM diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index b26bb8101981..c64778fd5e13 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -123,6 +123,48 @@ static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst) return ((struct rt6_info *)dst)->rt6i_idev; } +static inline void rt6_clean_expires(struct rt6_info *rt) +{ + if (!(rt->rt6i_flags & RTF_EXPIRES) && rt->dst.from) + dst_release(rt->dst.from); + + rt->rt6i_flags &= ~RTF_EXPIRES; + rt->dst.expires = 0; +} + +static inline void rt6_set_expires(struct rt6_info *rt, unsigned long expires) +{ + if (!(rt->rt6i_flags & RTF_EXPIRES) && rt->dst.from) + dst_release(rt->dst.from); + + rt->rt6i_flags |= RTF_EXPIRES; + rt->dst.expires = expires; +} + +static inline void rt6_update_expires(struct rt6_info *rt, int timeout) +{ + if (!(rt->rt6i_flags & RTF_EXPIRES) && rt->dst.from) + dst_release(rt->dst.from); + + dst_set_expires(&rt->dst, timeout); + rt->rt6i_flags |= RTF_EXPIRES; +} + +static inline void rt6_set_from(struct rt6_info *rt, struct rt6_info *from) +{ + struct dst_entry *new = (struct dst_entry *) from; + + if (!(rt->rt6i_flags & RTF_EXPIRES) && rt->dst.from) { + if (new == rt->dst.from) + return; + dst_release(rt->dst.from); + } + + rt->rt6i_flags &= ~RTF_EXPIRES; + rt->dst.from = new; + dst_hold(new); +} + struct fib6_walker_t { struct list_head lh; struct fib6_node *root, *node; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 6a3bb6077e19..7d5cb975cc6f 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -803,8 +803,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp) ip6_del_rt(rt); rt = NULL; } else if (!(rt->rt6i_flags & RTF_EXPIRES)) { - rt->dst.expires = expires; - rt->rt6i_flags |= RTF_EXPIRES; + rt6_set_expires(rt, expires); } } dst_release(&rt->dst); @@ -1887,11 +1886,9 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) rt = NULL; } else if (addrconf_finite_timeout(rt_expires)) { /* not infinity */ - rt->dst.expires = jiffies + rt_expires; - rt->rt6i_flags |= RTF_EXPIRES; + rt6_set_expires(rt, jiffies + rt_expires); } else { - rt->rt6i_flags &= ~RTF_EXPIRES; - rt->dst.expires = 0; + rt6_clean_expires(rt); } } else if (valid_lft) { clock_t expires = 0; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 5b27fbcae346..93717435013e 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -673,11 +673,10 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, &rt->rt6i_gateway)) { if (!(iter->rt6i_flags & RTF_EXPIRES)) return -EEXIST; - iter->dst.expires = rt->dst.expires; - if (!(rt->rt6i_flags & RTF_EXPIRES)) { - iter->rt6i_flags &= ~RTF_EXPIRES; - iter->dst.expires = 0; - } + if (!(rt->rt6i_flags & RTF_EXPIRES)) + rt6_clean_expires(iter); + else + rt6_set_expires(iter, rt->dst.expires); return -EEXIST; } } diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 3dcdb81ec3e8..176b469322ac 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1264,8 +1264,7 @@ static void ndisc_router_discovery(struct sk_buff *skb) } if (rt) - rt->dst.expires = jiffies + (HZ * lifetime); - + rt6_set_expires(rt, jiffies + (HZ * lifetime)); if (ra_msg->icmph.icmp6_hop_limit) { in6_dev->cnf.hop_limit = ra_msg->icmph.icmp6_hop_limit; if (rt) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 3992e26a6039..bc4888d902b2 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -62,7 +62,7 @@ #include #endif -static struct rt6_info *ip6_rt_copy(const struct rt6_info *ort, +static struct rt6_info *ip6_rt_copy(struct rt6_info *ort, const struct in6_addr *dest); static struct dst_entry *ip6_dst_check(struct dst_entry *dst, u32 cookie); static unsigned int ip6_default_advmss(const struct dst_entry *dst); @@ -285,6 +285,10 @@ static void ip6_dst_destroy(struct dst_entry *dst) rt->rt6i_idev = NULL; in6_dev_put(idev); } + + if (!(rt->rt6i_flags & RTF_EXPIRES) && dst->from) + dst_release(dst->from); + if (peer) { rt->rt6i_peer = NULL; inet_putpeer(peer); @@ -329,8 +333,17 @@ static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev, static __inline__ int rt6_check_expired(const struct rt6_info *rt) { - return (rt->rt6i_flags & RTF_EXPIRES) && - time_after(jiffies, rt->dst.expires); + struct rt6_info *ort = NULL; + + if (rt->rt6i_flags & RTF_EXPIRES) { + if (time_after(jiffies, rt->dst.expires)) + return 1; + } else if (rt->dst.from) { + ort = (struct rt6_info *) rt->dst.from; + return (ort->rt6i_flags & RTF_EXPIRES) && + time_after(jiffies, ort->dst.expires); + } + return 0; } static inline int rt6_need_strict(const struct in6_addr *daddr) @@ -620,12 +633,11 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, (rt->rt6i_flags & ~RTF_PREF_MASK) | RTF_PREF(pref); if (rt) { - if (!addrconf_finite_timeout(lifetime)) { - rt->rt6i_flags &= ~RTF_EXPIRES; - } else { - rt->dst.expires = jiffies + HZ * lifetime; - rt->rt6i_flags |= RTF_EXPIRES; - } + if (!addrconf_finite_timeout(lifetime)) + rt6_clean_expires(rt); + else + rt6_set_expires(rt, jiffies + HZ * lifetime); + dst_release(&rt->dst); } return 0; @@ -730,7 +742,7 @@ int ip6_ins_rt(struct rt6_info *rt) return __ip6_ins_rt(rt, &info); } -static struct rt6_info *rt6_alloc_cow(const struct rt6_info *ort, +static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort, const struct in6_addr *daddr, const struct in6_addr *saddr) { @@ -954,10 +966,10 @@ struct dst_entry *ip6_blackhole_route(struct net *net, struct dst_entry *dst_ori rt->rt6i_idev = ort->rt6i_idev; if (rt->rt6i_idev) in6_dev_hold(rt->rt6i_idev); - rt->dst.expires = 0; rt->rt6i_gateway = ort->rt6i_gateway; - rt->rt6i_flags = ort->rt6i_flags & ~RTF_EXPIRES; + rt->rt6i_flags = ort->rt6i_flags; + rt6_clean_expires(rt); rt->rt6i_metric = 0; memcpy(&rt->rt6i_dst, &ort->rt6i_dst, sizeof(struct rt6key)); @@ -1019,10 +1031,9 @@ static void ip6_link_failure(struct sk_buff *skb) rt = (struct rt6_info *) skb_dst(skb); if (rt) { - if (rt->rt6i_flags & RTF_CACHE) { - dst_set_expires(&rt->dst, 0); - rt->rt6i_flags |= RTF_EXPIRES; - } else if (rt->rt6i_node && (rt->rt6i_flags & RTF_DEFAULT)) + if (rt->rt6i_flags & RTF_CACHE) + rt6_update_expires(rt, 0); + else if (rt->rt6i_node && (rt->rt6i_flags & RTF_DEFAULT)) rt->rt6i_node->fn_sernum = -1; } } @@ -1289,9 +1300,12 @@ int ip6_route_add(struct fib6_config *cfg) } rt->dst.obsolete = -1; - rt->dst.expires = (cfg->fc_flags & RTF_EXPIRES) ? - jiffies + clock_t_to_jiffies(cfg->fc_expires) : - 0; + + if (cfg->fc_flags & RTF_EXPIRES) + rt6_set_expires(rt, jiffies + + clock_t_to_jiffies(cfg->fc_expires)); + else + rt6_clean_expires(rt); if (cfg->fc_protocol == RTPROT_UNSPEC) cfg->fc_protocol = RTPROT_BOOT; @@ -1736,8 +1750,8 @@ again: features |= RTAX_FEATURE_ALLFRAG; dst_metric_set(&rt->dst, RTAX_FEATURES, features); } - dst_set_expires(&rt->dst, net->ipv6.sysctl.ip6_rt_mtu_expires); - rt->rt6i_flags |= RTF_MODIFIED|RTF_EXPIRES; + rt6_update_expires(rt, net->ipv6.sysctl.ip6_rt_mtu_expires); + rt->rt6i_flags |= RTF_MODIFIED; goto out; } @@ -1765,9 +1779,8 @@ again: * which is 10 mins. After 10 mins the decreased pmtu is expired * and detecting PMTU increase will be automatically happened. */ - dst_set_expires(&nrt->dst, net->ipv6.sysctl.ip6_rt_mtu_expires); - nrt->rt6i_flags |= RTF_DYNAMIC|RTF_EXPIRES; - + rt6_update_expires(nrt, net->ipv6.sysctl.ip6_rt_mtu_expires); + nrt->rt6i_flags |= RTF_DYNAMIC; ip6_ins_rt(nrt); } out: @@ -1799,7 +1812,7 @@ void rt6_pmtu_discovery(const struct in6_addr *daddr, const struct in6_addr *sad * Misc support functions */ -static struct rt6_info *ip6_rt_copy(const struct rt6_info *ort, +static struct rt6_info *ip6_rt_copy(struct rt6_info *ort, const struct in6_addr *dest) { struct net *net = dev_net(ort->dst.dev); @@ -1819,10 +1832,14 @@ static struct rt6_info *ip6_rt_copy(const struct rt6_info *ort, if (rt->rt6i_idev) in6_dev_hold(rt->rt6i_idev); rt->dst.lastuse = jiffies; - rt->dst.expires = 0; rt->rt6i_gateway = ort->rt6i_gateway; - rt->rt6i_flags = ort->rt6i_flags & ~RTF_EXPIRES; + rt->rt6i_flags = ort->rt6i_flags; + if ((ort->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) == + (RTF_DEFAULT | RTF_ADDRCONF)) + rt6_set_from(rt, ort); + else + rt6_clean_expires(rt); rt->rt6i_metric = 0; #ifdef CONFIG_IPV6_SUBTREES From 9a5d2bd99e0dfe9a31b3c160073ac445ba3d773f Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Sun, 8 Apr 2012 10:01:44 +0000 Subject: [PATCH 02/11] ppp: Fix race condition with queue start/stop Commit e675f0cc9a872fd152edc0c77acfed19bf28b81e ("ppp: Don't stop and restart queue on every TX packet") introduced a race condition which could leave the net queue stopped even when the channel is no longer busy. By calling netif_stop_queue() from ppp_start_xmit(), based on the return value from ppp_xmit_process() but *after* all the locks have been dropped, we could potentially do so *after* the channel has actually finished transmitting and attempted to re-wake the queue. Fix this by moving the netif_stop_queue() into ppp_xmit_process() under the xmit lock. I hadn't done this previously, because it gets called from other places than ppp_start_xmit(). But I now think it's the better option. The net queue *should* be stopped if the channel becomes congested due to writes from pppd, anyway. Signed-off-by: David Woodhouse Signed-off-by: David S. Miller --- drivers/net/ppp/ppp_generic.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c index 33f8c51968b6..21d7151fb0ab 100644 --- a/drivers/net/ppp/ppp_generic.c +++ b/drivers/net/ppp/ppp_generic.c @@ -235,7 +235,7 @@ struct ppp_net { /* Prototypes. */ static int ppp_unattached_ioctl(struct net *net, struct ppp_file *pf, struct file *file, unsigned int cmd, unsigned long arg); -static int ppp_xmit_process(struct ppp *ppp); +static void ppp_xmit_process(struct ppp *ppp); static void ppp_send_frame(struct ppp *ppp, struct sk_buff *skb); static void ppp_push(struct ppp *ppp); static void ppp_channel_push(struct channel *pch); @@ -969,8 +969,7 @@ ppp_start_xmit(struct sk_buff *skb, struct net_device *dev) put_unaligned_be16(proto, pp); skb_queue_tail(&ppp->file.xq, skb); - if (!ppp_xmit_process(ppp)) - netif_stop_queue(dev); + ppp_xmit_process(ppp); return NETDEV_TX_OK; outf: @@ -1048,11 +1047,10 @@ static void ppp_setup(struct net_device *dev) * Called to do any work queued up on the transmit side * that can now be done. */ -static int +static void ppp_xmit_process(struct ppp *ppp) { struct sk_buff *skb; - int ret = 0; ppp_xmit_lock(ppp); if (!ppp->closing) { @@ -1062,13 +1060,12 @@ ppp_xmit_process(struct ppp *ppp) ppp_send_frame(ppp, skb); /* If there's no work left to do, tell the core net code that we can accept some more. */ - if (!ppp->xmit_pending && !skb_peek(&ppp->file.xq)) { + if (!ppp->xmit_pending && !skb_peek(&ppp->file.xq)) netif_wake_queue(ppp->dev); - ret = 1; - } + else + netif_stop_queue(ppp->dev); } ppp_xmit_unlock(ppp); - return ret; } static inline struct sk_buff * From ca8f4fb21d08747013cce9cf1840aa5bfc31f2d8 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 9 Apr 2012 00:24:02 +0000 Subject: [PATCH 03/11] skbuff: struct ubuf_info callback type safety The skb struct ubuf_info callback gets passed struct ubuf_info itself, not the arg value as the field name and the function signature seem to imply. Rename the arg field to ctx to match usage, add documentation and change the callback argument type to make usage clear and to have compiler check correctness. Signed-off-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- drivers/vhost/net.c | 2 +- drivers/vhost/vhost.c | 5 ++--- drivers/vhost/vhost.h | 2 +- include/linux/skbuff.h | 7 ++++--- 4 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index f0da2c32fbde..1f21d2a1e528 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -238,7 +238,7 @@ static void handle_tx(struct vhost_net *net) vq->heads[vq->upend_idx].len = len; ubuf->callback = vhost_zerocopy_callback; - ubuf->arg = vq->ubufs; + ubuf->ctx = vq->ubufs; ubuf->desc = vq->upend_idx; msg.msg_control = ubuf; msg.msg_controllen = sizeof(ubuf); diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index 947f00d8e091..51e4c1eeec4f 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -1598,10 +1598,9 @@ void vhost_ubuf_put_and_wait(struct vhost_ubuf_ref *ubufs) kfree(ubufs); } -void vhost_zerocopy_callback(void *arg) +void vhost_zerocopy_callback(struct ubuf_info *ubuf) { - struct ubuf_info *ubuf = arg; - struct vhost_ubuf_ref *ubufs = ubuf->arg; + struct vhost_ubuf_ref *ubufs = ubuf->ctx; struct vhost_virtqueue *vq = ubufs->vq; /* set len = 1 to mark this desc buffers done DMA */ diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h index 8dcf4cca6bf2..8de1fd5b8efb 100644 --- a/drivers/vhost/vhost.h +++ b/drivers/vhost/vhost.h @@ -188,7 +188,7 @@ bool vhost_enable_notify(struct vhost_dev *, struct vhost_virtqueue *); int vhost_log_write(struct vhost_virtqueue *vq, struct vhost_log *log, unsigned int log_num, u64 len); -void vhost_zerocopy_callback(void *arg); +void vhost_zerocopy_callback(struct ubuf_info *); int vhost_zerocopy_signal_used(struct vhost_virtqueue *vq); #define vq_err(vq, fmt, ...) do { \ diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 70a3f8d49118..775292a66fa4 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -238,11 +238,12 @@ enum { /* * The callback notifies userspace to release buffers when skb DMA is done in * lower device, the skb last reference should be 0 when calling this. - * The desc is used to track userspace buffer index. + * The ctx field is used to track device context. + * The desc field is used to track userspace buffer index. */ struct ubuf_info { - void (*callback)(void *); - void *arg; + void (*callback)(struct ubuf_info *); + void *ctx; unsigned long desc; }; From 03662e41c7cff64a776bfb1b3816de4be43de881 Mon Sep 17 00:00:00 2001 From: Tony Zelenoff Date: Wed, 11 Apr 2012 06:15:03 +0000 Subject: [PATCH 04/11] atl1: fix kernel panic in case of DMA errors Problem: There was two separate work_struct structures which share one handler. Unfortunately getting atl1_adapter structure from work_struct in case of DMA error was done from incorrect offset which cause kernel panics. Solution: The useless work_struct for DMA error removed and handler name changed to more generic one. Signed-off-by: Tony Zelenoff Signed-off-by: David S. Miller --- drivers/net/ethernet/atheros/atlx/atl1.c | 12 +++++------- drivers/net/ethernet/atheros/atlx/atl1.h | 3 +-- drivers/net/ethernet/atheros/atlx/atlx.c | 2 +- 3 files changed, 7 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/atheros/atlx/atl1.c b/drivers/net/ethernet/atheros/atlx/atl1.c index 40ac41436549..c926857e8205 100644 --- a/drivers/net/ethernet/atheros/atlx/atl1.c +++ b/drivers/net/ethernet/atheros/atlx/atl1.c @@ -2476,7 +2476,7 @@ static irqreturn_t atl1_intr(int irq, void *data) "pcie phy link down %x\n", status); if (netif_running(adapter->netdev)) { /* reset MAC */ iowrite32(0, adapter->hw.hw_addr + REG_IMR); - schedule_work(&adapter->pcie_dma_to_rst_task); + schedule_work(&adapter->reset_dev_task); return IRQ_HANDLED; } } @@ -2488,7 +2488,7 @@ static irqreturn_t atl1_intr(int irq, void *data) "pcie DMA r/w error (status = 0x%x)\n", status); iowrite32(0, adapter->hw.hw_addr + REG_IMR); - schedule_work(&adapter->pcie_dma_to_rst_task); + schedule_work(&adapter->reset_dev_task); return IRQ_HANDLED; } @@ -2633,10 +2633,10 @@ static void atl1_down(struct atl1_adapter *adapter) atl1_clean_rx_ring(adapter); } -static void atl1_tx_timeout_task(struct work_struct *work) +static void atl1_reset_dev_task(struct work_struct *work) { struct atl1_adapter *adapter = - container_of(work, struct atl1_adapter, tx_timeout_task); + container_of(work, struct atl1_adapter, reset_dev_task); struct net_device *netdev = adapter->netdev; netif_device_detach(netdev); @@ -3038,12 +3038,10 @@ static int __devinit atl1_probe(struct pci_dev *pdev, (unsigned long)adapter); adapter->phy_timer_pending = false; - INIT_WORK(&adapter->tx_timeout_task, atl1_tx_timeout_task); + INIT_WORK(&adapter->reset_dev_task, atl1_reset_dev_task); INIT_WORK(&adapter->link_chg_task, atlx_link_chg_task); - INIT_WORK(&adapter->pcie_dma_to_rst_task, atl1_tx_timeout_task); - err = register_netdev(netdev); if (err) goto err_common; diff --git a/drivers/net/ethernet/atheros/atlx/atl1.h b/drivers/net/ethernet/atheros/atlx/atl1.h index 109d6da8be97..e04bf4d71e46 100644 --- a/drivers/net/ethernet/atheros/atlx/atl1.h +++ b/drivers/net/ethernet/atheros/atlx/atl1.h @@ -758,9 +758,8 @@ struct atl1_adapter { u16 link_speed; u16 link_duplex; spinlock_t lock; - struct work_struct tx_timeout_task; + struct work_struct reset_dev_task; struct work_struct link_chg_task; - struct work_struct pcie_dma_to_rst_task; struct timer_list phy_config_timer; bool phy_timer_pending; diff --git a/drivers/net/ethernet/atheros/atlx/atlx.c b/drivers/net/ethernet/atheros/atlx/atlx.c index 3cd8837236dc..c9e9dc57986c 100644 --- a/drivers/net/ethernet/atheros/atlx/atlx.c +++ b/drivers/net/ethernet/atheros/atlx/atlx.c @@ -194,7 +194,7 @@ static void atlx_tx_timeout(struct net_device *netdev) { struct atlx_adapter *adapter = netdev_priv(netdev); /* Do the reset outside of interrupt context */ - schedule_work(&adapter->tx_timeout_task); + schedule_work(&adapter->reset_dev_task); } /* From a8c9cb106fe79c28d6b7f1397652cadd228715ff Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Wed, 11 Apr 2012 22:10:54 +0000 Subject: [PATCH 05/11] 8139cp: set intr mask after its handler is registered We set intr mask before its handler is registered, this does not work well when 8139cp is sharing irq line with other devices. As the irq could be enabled by the device before 8139cp's hander is registered which may lead unhandled irq. Fix this by introducing an helper cp_irq_enable() and call it after request_irq(). Signed-off-by: Jason Wang Reviewed-by: Flavio Leitner Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/8139cp.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index abc79076f867..b3287c0fe279 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -958,6 +958,11 @@ static inline void cp_start_hw (struct cp_private *cp) cpw8(Cmd, RxOn | TxOn); } +static void cp_enable_irq(struct cp_private *cp) +{ + cpw16_f(IntrMask, cp_intr_mask); +} + static void cp_init_hw (struct cp_private *cp) { struct net_device *dev = cp->dev; @@ -997,8 +1002,6 @@ static void cp_init_hw (struct cp_private *cp) cpw16(MultiIntr, 0); - cpw16_f(IntrMask, cp_intr_mask); - cpw8_f(Cfg9346, Cfg9346_Lock); } @@ -1130,6 +1133,8 @@ static int cp_open (struct net_device *dev) if (rc) goto err_out_hw; + cp_enable_irq(cp); + netif_carrier_off(dev); mii_check_media(&cp->mii_if, netif_msg_link(cp), true); netif_start_queue(dev); @@ -2031,6 +2036,7 @@ static int cp_resume (struct pci_dev *pdev) /* FIXME: sh*t may happen if the Rx ring buffer is depleted */ cp_init_rings_index (cp); cp_init_hw (cp); + cp_enable_irq(cp); netif_start_queue (dev); spin_lock_irqsave (&cp->lock, flags); From 59a54f3082b9847c1240e24879750e3a03f96754 Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Thu, 12 Apr 2012 01:11:12 +0000 Subject: [PATCH 06/11] drivers/net/ethernet/xilinx/axi ethernet: Correct Copyright Also fix MAINTAINERS file to reflect autorship. Daniel and Ariane changed coding style but not any functional changes in the driver itself. Signed-off-by: Michal Simek Signed-off-by: David S. Miller --- MAINTAINERS | 4 ++-- drivers/net/ethernet/xilinx/xilinx_axienet.h | 4 +--- drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 6 +++--- drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c | 6 +++--- 4 files changed, 9 insertions(+), 11 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 32671e00800d..7cef7be8f840 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7577,8 +7577,8 @@ F: Documentation/filesystems/xfs.txt F: fs/xfs/ XILINX AXI ETHERNET DRIVER -M: Ariane Keller -M: Daniel Borkmann +M: Anirudha Sarangi +M: John Linn S: Maintained F: drivers/net/ethernet/xilinx/xilinx_axienet* diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet.h b/drivers/net/ethernet/xilinx/xilinx_axienet.h index cc83af083fd7..44b8d2bad8c3 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet.h +++ b/drivers/net/ethernet/xilinx/xilinx_axienet.h @@ -2,9 +2,7 @@ * Definitions for Xilinx Axi Ethernet device driver. * * Copyright (c) 2009 Secret Lab Technologies, Ltd. - * Copyright (c) 2010 Xilinx, Inc. All rights reserved. - * Copyright (c) 2012 Daniel Borkmann, - * Copyright (c) 2012 Ariane Keller, + * Copyright (c) 2010 - 2012 Xilinx, Inc. All rights reserved. */ #ifndef XILINX_AXIENET_H diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c index 2fcbeba6814b..9c365e192a31 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c @@ -4,9 +4,9 @@ * Copyright (c) 2008 Nissin Systems Co., Ltd., Yoshio Kashiwagi * Copyright (c) 2005-2008 DLA Systems, David H. Lynch Jr. * Copyright (c) 2008-2009 Secret Lab Technologies Ltd. - * Copyright (c) 2010 Xilinx, Inc. All rights reserved. - * Copyright (c) 2012 Daniel Borkmann, - * Copyright (c) 2012 Ariane Keller, + * Copyright (c) 2010 - 2011 Michal Simek + * Copyright (c) 2010 - 2011 PetaLogix + * Copyright (c) 2010 - 2012 Xilinx, Inc. All rights reserved. * * This is a driver for the Xilinx Axi Ethernet which is used in the Virtex6 * and Spartan6. diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c b/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c index d70b6e79f6c0..e90e1f46121e 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c @@ -2,9 +2,9 @@ * MDIO bus driver for the Xilinx Axi Ethernet device * * Copyright (c) 2009 Secret Lab Technologies, Ltd. - * Copyright (c) 2010 Xilinx, Inc. All rights reserved. - * Copyright (c) 2012 Daniel Borkmann, - * Copyright (c) 2012 Ariane Keller, + * Copyright (c) 2010 - 2011 Michal Simek + * Copyright (c) 2010 - 2011 PetaLogix + * Copyright (c) 2010 - 2012 Xilinx, Inc. All rights reserved. */ #include From 51c61a2838c33dab7b6659b9a3e008bb1b40bc9b Mon Sep 17 00:00:00 2001 From: Matt Renzelmann Date: Fri, 13 Apr 2012 07:59:40 +0000 Subject: [PATCH 07/11] ks8851: Fix missing mutex_lock/unlock Move the ks8851_rdreg16 call above the call to request_irq and cache the result for subsequent repeated use. A spurious interrupt may otherwise cause a crash. Thanks to Stephen Boyd, Flavio Leitner, and Ben Hutchings for feedback. Signed-off-by: Matt Renzelmann Signed-off-by: David S. Miller --- drivers/net/ethernet/micrel/ks8851.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/micrel/ks8851.c b/drivers/net/ethernet/micrel/ks8851.c index c722aa607d07..e5dc0757f077 100644 --- a/drivers/net/ethernet/micrel/ks8851.c +++ b/drivers/net/ethernet/micrel/ks8851.c @@ -1418,6 +1418,7 @@ static int __devinit ks8851_probe(struct spi_device *spi) struct net_device *ndev; struct ks8851_net *ks; int ret; + unsigned cider; ndev = alloc_etherdev(sizeof(struct ks8851_net)); if (!ndev) @@ -1484,8 +1485,8 @@ static int __devinit ks8851_probe(struct spi_device *spi) ks8851_soft_reset(ks, GRR_GSR); /* simple check for a valid chip being connected to the bus */ - - if ((ks8851_rdreg16(ks, KS_CIDER) & ~CIDER_REV_MASK) != CIDER_ID) { + cider = ks8851_rdreg16(ks, KS_CIDER); + if ((cider & ~CIDER_REV_MASK) != CIDER_ID) { dev_err(&spi->dev, "failed to read device ID\n"); ret = -ENODEV; goto err_id; @@ -1516,8 +1517,7 @@ static int __devinit ks8851_probe(struct spi_device *spi) } netdev_info(ndev, "revision %d, MAC %pM, IRQ %d, %s EEPROM\n", - CIDER_REV_GET(ks8851_rdreg16(ks, KS_CIDER)), - ndev->dev_addr, ndev->irq, + CIDER_REV_GET(cider), ndev->dev_addr, ndev->irq, ks->rc_ccr & CCR_EEPROM ? "has" : "no"); return 0; From 3c5e979bd037888dd7d722da22da4b43659af485 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 12 Apr 2012 05:54:09 +0000 Subject: [PATCH 08/11] net: smsc911x: fix skb handling in receive path The SMSC911x driver resets the ->head, ->data and ->tail pointers in the skb on the reset path in order to avoid buffer overflow due to packet padding performed by the hardware. This patch fixes the receive path so that the skb pointers are fixed up after the data has been read from the device, The error path is also fixed to use number of words consistently and prevent erroneous FIFO fastforwarding when skipping over bad data. Signed-off-by: Will Deacon Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/ethernet/smsc/smsc911x.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index 4a6971027076..5aa2dbe2bfb5 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1166,10 +1166,8 @@ smsc911x_rx_counterrors(struct net_device *dev, unsigned int rxstat) /* Quickly dumps bad packets */ static void -smsc911x_rx_fastforward(struct smsc911x_data *pdata, unsigned int pktbytes) +smsc911x_rx_fastforward(struct smsc911x_data *pdata, unsigned int pktwords) { - unsigned int pktwords = (pktbytes + NET_IP_ALIGN + 3) >> 2; - if (likely(pktwords >= 4)) { unsigned int timeout = 500; unsigned int val; @@ -1233,7 +1231,7 @@ static int smsc911x_poll(struct napi_struct *napi, int budget) continue; } - skb = netdev_alloc_skb(dev, pktlength + NET_IP_ALIGN); + skb = netdev_alloc_skb(dev, pktwords << 2); if (unlikely(!skb)) { SMSC_WARN(pdata, rx_err, "Unable to allocate skb for rx packet"); @@ -1243,14 +1241,12 @@ static int smsc911x_poll(struct napi_struct *napi, int budget) break; } - skb->data = skb->head; - skb_reset_tail_pointer(skb); + pdata->ops->rx_readfifo(pdata, + (unsigned int *)skb->data, pktwords); /* Align IP on 16B boundary */ skb_reserve(skb, NET_IP_ALIGN); skb_put(skb, pktlength - 4); - pdata->ops->rx_readfifo(pdata, - (unsigned int *)skb->head, pktwords); skb->protocol = eth_type_trans(skb, dev); skb_checksum_none_assert(skb); netif_receive_skb(skb); @@ -1565,7 +1561,7 @@ static int smsc911x_open(struct net_device *dev) smsc911x_reg_write(pdata, FIFO_INT, temp); /* set RX Data offset to 2 bytes for alignment */ - smsc911x_reg_write(pdata, RX_CFG, (2 << 8)); + smsc911x_reg_write(pdata, RX_CFG, (NET_IP_ALIGN << 8)); /* enable NAPI polling before enabling RX interrupts */ napi_enable(&pdata->napi); From 92fe173391b3e0d0a7212fa8b9d72c8d61f31c26 Mon Sep 17 00:00:00 2001 From: Bruce Allan Date: Thu, 12 Apr 2012 06:27:03 +0000 Subject: [PATCH 09/11] e1000e: issues in Sx on 82577/8/9 A workaround was previously put in the driver to reset the device when transitioning to Sx in order to activate the changed settings of the PHY OEM bits (Low Power Link Up, or LPLU, and GbE disable configuration) for 82577/8/9 devices. After further review, it was found such a reset can cause the 82579 to confuse which version of 82579 it actually is and broke LPLU on all 82577/8/9 devices. The workaround during an S0->Sx transition on 82579 (instead of resetting the PHY) is to restart auto-negotiation after the OEM bits are configured; the restart of auto-negotiation activates the new OEM bits as does the reset. With 82577/8, the reset is changed to a generic reset which fixes the LPLU bits getting set wrong. Signed-off-by: Bruce Allan Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/e1000e/ich8lan.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c index 64c76443a7aa..b461c24945e3 100644 --- a/drivers/net/ethernet/intel/e1000e/ich8lan.c +++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c @@ -1310,10 +1310,6 @@ static s32 e1000_oem_bits_config_ich8lan(struct e1000_hw *hw, bool d0_state) if (mac_reg & E1000_PHY_CTRL_D0A_LPLU) oem_reg |= HV_OEM_BITS_LPLU; - - /* Set Restart auto-neg to activate the bits */ - if (!hw->phy.ops.check_reset_block(hw)) - oem_reg |= HV_OEM_BITS_RESTART_AN; } else { if (mac_reg & (E1000_PHY_CTRL_GBE_DISABLE | E1000_PHY_CTRL_NOND0A_GBE_DISABLE)) @@ -1324,6 +1320,11 @@ static s32 e1000_oem_bits_config_ich8lan(struct e1000_hw *hw, bool d0_state) oem_reg |= HV_OEM_BITS_LPLU; } + /* Set Restart auto-neg to activate the bits */ + if ((d0_state || (hw->mac.type != e1000_pchlan)) && + !hw->phy.ops.check_reset_block(hw)) + oem_reg |= HV_OEM_BITS_RESTART_AN; + ret_val = hw->phy.ops.write_reg_locked(hw, HV_OEM_BITS, oem_reg); release: @@ -3682,7 +3683,11 @@ void e1000_suspend_workarounds_ich8lan(struct e1000_hw *hw) if (hw->mac.type >= e1000_pchlan) { e1000_oem_bits_config_ich8lan(hw, false); - e1000_phy_hw_reset_ich8lan(hw); + + /* Reset PHY to activate OEM bits on 82577/8 */ + if (hw->mac.type == e1000_pchlan) + e1000e_phy_hw_reset_generic(hw); + ret_val = hw->phy.ops.acquire(hw); if (ret_val) return; From c509e754af96882eb46130e215cc4aa8f763d4ac Mon Sep 17 00:00:00 2001 From: Don Skidmore Date: Thu, 5 Apr 2012 08:12:05 +0000 Subject: [PATCH 10/11] ixgbe: fix WoL issue with fiber There are times we turn of the laser before shutdown. This is a bad thing if we want to wake on lan to work so now we make sure the laser is on before shutdown if we support WoL. Signed-off-by: Don Skidmore Tested-by: Stephen Ko Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 3e26b1f9ac75..dac7c01f8332 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -4893,6 +4893,16 @@ static int __ixgbe_shutdown(struct pci_dev *pdev, bool *enable_wake) if (wufc) { ixgbe_set_rx_mode(netdev); + /* + * enable the optics for both mult-speed fiber and + * 82599 SFP+ fiber as we can WoL. + */ + if (hw->mac.ops.enable_tx_laser && + (hw->phy.multispeed_fiber || + (hw->mac.ops.get_media_type(hw) == ixgbe_media_type_fiber && + hw->mac.type == ixgbe_mac_82599EB))) + hw->mac.ops.enable_tx_laser(hw); + /* turn on all-multi mode if wake on multicast is enabled */ if (wufc & IXGBE_WUFC_MC) { fctrl = IXGBE_READ_REG(hw, IXGBE_FCTRL); From 8a9a0ea6032186e3030419262678d652b88bf6a8 Mon Sep 17 00:00:00 2001 From: Davide Ciminaghi Date: Fri, 13 Apr 2012 04:48:25 +0000 Subject: [PATCH 11/11] net/ethernet: ks8851_mll fix rx frame buffer overflow At the beginning of ks_rcv(), a for loop retrieves the header information relevant to all the frames stored in the mac's internal buffers. The number of pending frames is stored as an 8 bits field in KS_RXFCTR. If interrupts are disabled long enough to allow for more than 32 frames to accumulate in the MAC's internal buffers, a buffer overflow occurs. This patch fixes the problem by making the driver's frame_head_info buffer big enough. Well actually, since the chip appears to have 12K of internal rx buffers and the shortest ethernet frame should be 64 bytes long, maybe the limit could be set to 12*1024/64 = 192 frames, but 255 should be safer. Signed-off-by: Davide Ciminaghi Signed-off-by: Raffaele Recalcati Signed-off-by: David S. Miller --- drivers/net/ethernet/micrel/ks8851_mll.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/micrel/ks8851_mll.c b/drivers/net/ethernet/micrel/ks8851_mll.c index b8104d9f4081..5ffde23ac8fb 100644 --- a/drivers/net/ethernet/micrel/ks8851_mll.c +++ b/drivers/net/ethernet/micrel/ks8851_mll.c @@ -40,7 +40,7 @@ #define DRV_NAME "ks8851_mll" static u8 KS_DEFAULT_MAC_ADDRESS[] = { 0x00, 0x10, 0xA1, 0x86, 0x95, 0x11 }; -#define MAX_RECV_FRAMES 32 +#define MAX_RECV_FRAMES 255 #define MAX_BUF_SIZE 2048 #define TX_BUF_SIZE 2000 #define RX_BUF_SIZE 2000