From: Antonio Quartulli <ordex@autistici.org> To: davem@davemloft.net Cc: netdev@vger.kernel.org, b.a.t.m.a.n@lists.open-mesh.org, Sven Eckelmann <sven@narfation.org>, Antonio Quartulli <ordex@autistici.org> Subject: [PATCH 06/16] batman-adv: Set special lockdep classes to avoid lockdep warning Date: Sun, 28 Oct 2012 12:17:07 +0100 [thread overview] Message-ID: <1351423037-5292-7-git-send-email-ordex@autistici.org> (raw) In-Reply-To: <1351423037-5292-1-git-send-email-ordex@autistici.org> From: Sven Eckelmann <sven@narfation.org> Transmissions over batman-adv devices always start another nested transmission over devices attached to the batman-adv interface. These devices usually use the ethernet lockdep class for the tx_queue lock which is also set by default for all batman-adv devices. Lockdep will detect a nested locking attempt of two locks with the same class and warn about a possible deadlock. This is the default and expected behavior and should not alarm the locking correctness prove mechanism. Therefore, the locks for all netdevice specific tx queues get a special batman-adv lock class to avoid a false positive for each transmission. Reported-by: Linus Luessing <linus.luessing@web.de> Signed-off-by: Sven Eckelmann <sven@narfation.org> Signed-off-by: Antonio Quartulli <ordex@autistici.org> --- net/batman-adv/soft-interface.c | 44 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index b9a28d2..948860a 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -347,7 +347,51 @@ out: return; } +/* batman-adv network devices have devices nesting below it and are a special + * "super class" of normal network devices; split their locks off into a + * separate class since they always nest. + */ +static struct lock_class_key batadv_netdev_xmit_lock_key; +static struct lock_class_key batadv_netdev_addr_lock_key; + +/** + * batadv_set_lockdep_class_one - Set lockdep class for a single tx queue + * @dev: device which owns the tx queue + * @txq: tx queue to modify + * @_unused: always NULL + */ +static void batadv_set_lockdep_class_one(struct net_device *dev, + struct netdev_queue *txq, + void *_unused) +{ + lockdep_set_class(&txq->_xmit_lock, &batadv_netdev_xmit_lock_key); +} + +/** + * batadv_set_lockdep_class - Set txq and addr_list lockdep class + * @dev: network device to modify + */ +static void batadv_set_lockdep_class(struct net_device *dev) +{ + lockdep_set_class(&dev->addr_list_lock, &batadv_netdev_addr_lock_key); + netdev_for_each_tx_queue(dev, batadv_set_lockdep_class_one, NULL); +} + +/** + * batadv_softif_init - Late stage initialization of soft interface + * @dev: registered network device to modify + * + * Returns error code on failures + */ +static int batadv_softif_init(struct net_device *dev) +{ + batadv_set_lockdep_class(dev); + + return 0; +} + static const struct net_device_ops batadv_netdev_ops = { + .ndo_init = batadv_softif_init, .ndo_open = batadv_interface_open, .ndo_stop = batadv_interface_release, .ndo_get_stats = batadv_interface_stats, -- 1.7.12.4
WARNING: multiple messages have this Message-ID (diff)
From: Antonio Quartulli <ordex@autistici.org> To: davem@davemloft.net Cc: netdev@vger.kernel.org, b.a.t.m.a.n@lists.open-mesh.org Subject: [B.A.T.M.A.N.] [PATCH 06/16] batman-adv: Set special lockdep classes to avoid lockdep warning Date: Sun, 28 Oct 2012 12:17:07 +0100 [thread overview] Message-ID: <1351423037-5292-7-git-send-email-ordex@autistici.org> (raw) In-Reply-To: <1351423037-5292-1-git-send-email-ordex@autistici.org> From: Sven Eckelmann <sven@narfation.org> Transmissions over batman-adv devices always start another nested transmission over devices attached to the batman-adv interface. These devices usually use the ethernet lockdep class for the tx_queue lock which is also set by default for all batman-adv devices. Lockdep will detect a nested locking attempt of two locks with the same class and warn about a possible deadlock. This is the default and expected behavior and should not alarm the locking correctness prove mechanism. Therefore, the locks for all netdevice specific tx queues get a special batman-adv lock class to avoid a false positive for each transmission. Reported-by: Linus Luessing <linus.luessing@web.de> Signed-off-by: Sven Eckelmann <sven@narfation.org> Signed-off-by: Antonio Quartulli <ordex@autistici.org> --- net/batman-adv/soft-interface.c | 44 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index b9a28d2..948860a 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -347,7 +347,51 @@ out: return; } +/* batman-adv network devices have devices nesting below it and are a special + * "super class" of normal network devices; split their locks off into a + * separate class since they always nest. + */ +static struct lock_class_key batadv_netdev_xmit_lock_key; +static struct lock_class_key batadv_netdev_addr_lock_key; + +/** + * batadv_set_lockdep_class_one - Set lockdep class for a single tx queue + * @dev: device which owns the tx queue + * @txq: tx queue to modify + * @_unused: always NULL + */ +static void batadv_set_lockdep_class_one(struct net_device *dev, + struct netdev_queue *txq, + void *_unused) +{ + lockdep_set_class(&txq->_xmit_lock, &batadv_netdev_xmit_lock_key); +} + +/** + * batadv_set_lockdep_class - Set txq and addr_list lockdep class + * @dev: network device to modify + */ +static void batadv_set_lockdep_class(struct net_device *dev) +{ + lockdep_set_class(&dev->addr_list_lock, &batadv_netdev_addr_lock_key); + netdev_for_each_tx_queue(dev, batadv_set_lockdep_class_one, NULL); +} + +/** + * batadv_softif_init - Late stage initialization of soft interface + * @dev: registered network device to modify + * + * Returns error code on failures + */ +static int batadv_softif_init(struct net_device *dev) +{ + batadv_set_lockdep_class(dev); + + return 0; +} + static const struct net_device_ops batadv_netdev_ops = { + .ndo_init = batadv_softif_init, .ndo_open = batadv_interface_open, .ndo_stop = batadv_interface_release, .ndo_get_stats = batadv_interface_stats, -- 1.7.12.4
next prev parent reply other threads:[~2012-10-28 11:18 UTC|newest] Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top 2012-10-28 11:17 pull request: batman-adv 2012-10-28 Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 01/16] batman-adv: use check_unicast_packet() in recv_roam_adv() Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 02/16] batman-adv: split hard_iface struct for each routing protocol Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 03/16] batman-adv: separate BATADV_ATTR_HIF_UINT sysfs name from hard-iface variable Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 22:59 ` David Miller 2012-10-28 22:59 ` [B.A.T.M.A.N.] " David Miller [not found] ` <20121028.185927.374901219739099234.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org> 2012-10-29 8:13 ` Antonio Quartulli 2012-10-29 8:13 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 04/16] batman-adv: consolidate duplicated primary_if checking code Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 05/16] batman-adv: return proper value in case of hash_add failure Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` Antonio Quartulli [this message] 2012-10-28 11:17 ` [B.A.T.M.A.N.] [PATCH 06/16] batman-adv: Set special lockdep classes to avoid lockdep warning Antonio Quartulli 2012-10-28 11:17 ` [PATCH 07/16] batman-adv: Remove extra check in batadv_bit_get_packet Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 08/16] batman-adv: Check return value of try_module_get Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 09/16] batman-adv: Only increase refcounter once for alternate router Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 10/16] batman-adv: don't allow ECTP traffic on batman-adv Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 11/16] batman-adv: properly store the roaming time Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 12/16] batman-adv: print packets re-routing on DBG_TT and ratelimit it Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 13/16] batman-adv: check for more space before accessing the skb Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 14/16] batman-adv: properly convert flag into a boolean value Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 15/16] batman-adv: pass the WIFI flag from the local to global entry Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli 2012-10-28 11:17 ` [PATCH 16/16] batman-adv: add kernel-doc for enum batadv_dbg_level Antonio Quartulli 2012-10-28 11:17 ` [B.A.T.M.A.N.] " Antonio Quartulli [not found] ` <1351423037-5292-17-git-send-email-ordex-GaUfNO9RBHfsrOwW+9ziJQ@public.gmane.org> 2012-10-28 13:32 ` Walter Robert Ditzler 2012-10-28 13:32 ` [B.A.T.M.A.N.] " Walter Robert Ditzler 2012-10-28 14:33 ` Sven Eckelmann 2012-10-29 8:58 pull request: batman-adv 202-10-29 Antonio Quartulli 2012-10-29 8:58 ` [PATCH 06/16] batman-adv: Set special lockdep classes to avoid lockdep warning Antonio Quartulli
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=1351423037-5292-7-git-send-email-ordex@autistici.org \ --to=ordex@autistici.org \ --cc=b.a.t.m.a.n@lists.open-mesh.org \ --cc=davem@davemloft.net \ --cc=netdev@vger.kernel.org \ --cc=sven@narfation.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.