From b8a1b1ce14252b59b2d5c89de25b54f9bfd4cc5e Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Wed, 14 Jan 2009 14:55:41 -0800 Subject: IPoIB: Fix hang in napi_disable() if P_Key is never found After commit fe25c561 ("IPoIB: Don't enable NAPI when it's already enabled"), if an interface is brought up but the corresponding P_Key never appears, then ipoib_stop() will hang in napi_disable(), because ipoib_open() returns before it does napi_enable(). Fix this by changing ipoib_open() to call napi_enable() even if the P_Key isn't present. Reported-by: Yossi Etigin Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/ipoib/ipoib_main.c | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) (limited to 'drivers/infiniband/ulp') diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index dce0443f9d6..0bd2a4ff084 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -106,23 +106,17 @@ int ipoib_open(struct net_device *dev) ipoib_dbg(priv, "bringing up interface\n"); - set_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags); + if (!test_and_set_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) + napi_enable(&priv->napi); if (ipoib_pkey_dev_delay_open(dev)) return 0; - napi_enable(&priv->napi); + if (ipoib_ib_dev_open(dev)) + goto err_disable; - if (ipoib_ib_dev_open(dev)) { - napi_disable(&priv->napi); - return -EINVAL; - } - - if (ipoib_ib_dev_up(dev)) { - ipoib_ib_dev_stop(dev, 1); - napi_disable(&priv->napi); - return -EINVAL; - } + if (ipoib_ib_dev_up(dev)) + goto err_stop; if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) { struct ipoib_dev_priv *cpriv; @@ -144,6 +138,15 @@ int ipoib_open(struct net_device *dev) netif_start_queue(dev); return 0; + +err_stop: + ipoib_ib_dev_stop(dev, 1); + +err_disable: + napi_disable(&priv->napi); + clear_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags); + + return -EINVAL; } static int ipoib_stop(struct net_device *dev) -- cgit v1.2.3-18-g5258 From cbbe1efa4972350286b52cb48aefaa11e198c0fb Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Wed, 14 Jan 2009 21:44:39 -0800 Subject: IPoIB: Fix deadlock between ipoib_open() and child interface create Fix a deadlock between child interface creation/deletion and ipoib start/stop. The former takes vlan_mutex, and then might take RTNL via register_netdev()/unregister_netdev(). The latter is executed with RTNL held, and tries to take vlan_mutex, which can lead to an AB-BA deadlock. Fix this by having the child interface creation/deletion code take the RTNL first so vlan_mutex always nests inside RTNL. We can use register_netdevice() for child interfaces because we form the interface name from the parent interface and hence don't need the '%' expansion of register_netdev(). Reported-by: Yossi Etigin Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/ipoib/ipoib_vlan.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'drivers/infiniband/ulp') diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c index 2cf1a408871..5a76a551035 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c @@ -61,6 +61,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) ppriv = netdev_priv(pdev); + rtnl_lock(); mutex_lock(&ppriv->vlan_mutex); /* @@ -111,7 +112,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) goto device_init_failed; } - result = register_netdev(priv->dev); + result = register_netdevice(priv->dev); if (result) { ipoib_warn(priv, "failed to initialize; error %i", result); goto register_failed; @@ -134,12 +135,13 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) list_add_tail(&priv->list, &ppriv->child_intfs); mutex_unlock(&ppriv->vlan_mutex); + rtnl_unlock(); return 0; sysfs_failed: ipoib_delete_debug_files(priv->dev); - unregister_netdev(priv->dev); + unregister_netdevice(priv->dev); register_failed: ipoib_dev_cleanup(priv->dev); @@ -149,6 +151,7 @@ device_init_failed: err: mutex_unlock(&ppriv->vlan_mutex); + rtnl_unlock(); return result; } @@ -162,10 +165,11 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey) ppriv = netdev_priv(pdev); + rtnl_lock(); mutex_lock(&ppriv->vlan_mutex); list_for_each_entry_safe(priv, tpriv, &ppriv->child_intfs, list) { if (priv->pkey == pkey) { - unregister_netdev(priv->dev); + unregister_netdevice(priv->dev); ipoib_dev_cleanup(priv->dev); list_del(&priv->list); free_netdev(priv->dev); @@ -175,6 +179,7 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey) } } mutex_unlock(&ppriv->vlan_mutex); + rtnl_unlock(); return ret; } -- cgit v1.2.3-18-g5258 From 3c20962086b0ceb5498ba840e5a91bf4a692aae9 Mon Sep 17 00:00:00 2001 From: Yossi Etigin Date: Fri, 16 Jan 2009 13:42:59 -0800 Subject: IPoIB: Do not print error messages for multicast join retries When IPoIB tries to join a multicast group, and the SA module's SM address handle is NULL (because of an SM change, etc), the join returns with -EAGAIN status. In that case, don't print an error message unless multicast debugging is enabled. Signed-off-by: Yossi Etigin Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/infiniband/ulp') diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 59d02e0b8df..425e31112ed 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -409,7 +409,7 @@ static int ipoib_mcast_join_complete(int status, } if (mcast->logcount++ < 20) { - if (status == -ETIMEDOUT) { + if (status == -ETIMEDOUT || status == -EAGAIN) { ipoib_dbg_mcast(priv, "multicast join failed for %pI6, status %d\n", mcast->mcmember.mgid.raw, status); } else { -- cgit v1.2.3-18-g5258