* [B.A.T.M.A.N.] [RFC] batman-adv: switch routing algorithm at runtime @ 2011-11-28 16:27 Marek Lindner 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change " Marek Lindner 2011-11-28 18:47 ` [B.A.T.M.A.N.] [RFC] batman-adv: switch " Andrew Lunn 0 siblings, 2 replies; 19+ messages in thread From: Marek Lindner @ 2011-11-28 16:27 UTC (permalink / raw) To: b.a.t.m.a.n Hi folks, after the previous discussions how to proceed with BATMAN V we came to the conclusion that a runtime switch for changing the routing algorithms would be the best solution. Hence, I drafted a couple of patches that I'd like to get some feedback on. Note, that these patches are not ready to be included. For instance, the rcu locking isn't clean. I'd like to focus your attention towards the following items: * How do we design the algorithm registration. At the moment batman_init() calls bat_iv_init() and possibly other algorithms in the future but this is less than ideal. I had hoped to find a mechanism which allows an init function to be declared and called inside of the respective routing algorithm files. These files can be compiled into the module or not. * The API and its implementation. It is highly likely that this API won't survive for very long as it has been designed to satisfy one routing algorithm only. However, changing it in the future should be easy. * Other ideas / comments ? Cheers, Marek ^ permalink raw reply [flat|nested] 19+ messages in thread
* [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change routing algorithm at runtime 2011-11-28 16:27 [B.A.T.M.A.N.] [RFC] batman-adv: switch routing algorithm at runtime Marek Lindner @ 2011-11-28 16:28 ` Marek Lindner 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure Marek Lindner ` (2 more replies) 2011-11-28 18:47 ` [B.A.T.M.A.N.] [RFC] batman-adv: switch " Andrew Lunn 1 sibling, 3 replies; 19+ messages in thread From: Marek Lindner @ 2011-11-28 16:28 UTC (permalink / raw) To: b.a.t.m.a.n; +Cc: Marek Lindner Signed-off-by: Marek Lindner <lindner_marek@yahoo.de> --- bat_debugfs.c | 8 ++++++ bat_iv_ogm.c | 9 ++++++ main.c | 75 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ main.h | 5 +++ soft-interface.c | 4 +++ types.h | 7 +++++ 6 files changed, 108 insertions(+), 0 deletions(-) diff --git a/bat_debugfs.c b/bat_debugfs.c index d0af9bf..05e2f1f 100644 --- a/bat_debugfs.c +++ b/bat_debugfs.c @@ -221,6 +221,12 @@ static void debug_log_cleanup(struct bat_priv *bat_priv) } #endif +static int bat_algorithms_open(struct inode *inode, struct file *file) +{ + struct net_device *net_dev = (struct net_device *)inode->i_private; + return single_open(file, bat_algo_seq_print_text, net_dev); +} + static int originators_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; @@ -274,6 +280,7 @@ struct bat_debuginfo bat_debuginfo_##_name = { \ } \ }; +static BAT_DEBUGINFO(routing_algos, S_IRUGO, bat_algorithms_open); static BAT_DEBUGINFO(originators, S_IRUGO, originators_open); static BAT_DEBUGINFO(gateways, S_IRUGO, gateways_open); static BAT_DEBUGINFO(softif_neigh, S_IRUGO, softif_neigh_open); @@ -282,6 +289,7 @@ static BAT_DEBUGINFO(transtable_local, S_IRUGO, transtable_local_open); static BAT_DEBUGINFO(vis_data, S_IRUGO, vis_data_open); static struct bat_debuginfo *mesh_debuginfos[] = { + &bat_debuginfo_routing_algos, &bat_debuginfo_originators, &bat_debuginfo_gateways, &bat_debuginfo_softif_neigh, diff --git a/bat_iv_ogm.c b/bat_iv_ogm.c index d60e1ba..a2b25ad 100644 --- a/bat_iv_ogm.c +++ b/bat_iv_ogm.c @@ -1169,3 +1169,12 @@ void bat_ogm_receive(const struct ethhdr *ethhdr, unsigned char *packet_buff, } while (bat_ogm_aggr_packet(buff_pos, packet_len, batman_ogm_packet->tt_num_changes)); } + +static struct bat_algo bat_algo_iv = { + .name = "BATMAN IV", +}; + +int __init bat_iv_init(void) +{ + return bat_algo_register(&bat_algo_iv); +} diff --git a/main.c b/main.c index fb87bdc..05b85f0 100644 --- a/main.c +++ b/main.c @@ -37,6 +37,7 @@ /* List manipulations on hardif_list have to be rtnl_lock()'ed, * list traversals just rcu-locked */ struct list_head hardif_list; +static struct hlist_head bat_algo_list; unsigned char broadcast_addr[] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; @@ -45,6 +46,9 @@ struct workqueue_struct *bat_event_workqueue; static int __init batman_init(void) { INIT_LIST_HEAD(&hardif_list); + INIT_HLIST_HEAD(&bat_algo_list); + + bat_iv_init(); /* the name should not be longer than 10 chars - see * http://lwn.net/Articles/23634/ */ @@ -173,6 +177,77 @@ int is_my_mac(const uint8_t *addr) } +static struct bat_algo *bat_algo_get(char *name) +{ + struct bat_algo *bat_algo = NULL, *bat_algo_tmp; + struct hlist_node *node; + + rcu_read_lock(); + hlist_for_each_entry_rcu(bat_algo_tmp, node, &bat_algo_list, list) { + if (strcmp(bat_algo_tmp->name, name) != 0) + continue; + + bat_algo = bat_algo_tmp; + break; + } + rcu_read_unlock(); + + return bat_algo; +} + +int bat_algo_register(struct bat_algo *bat_algo) +{ + struct bat_algo *bat_algo_tmp; + int ret = -1; + + bat_algo_tmp = bat_algo_get(bat_algo->name); + if (bat_algo_tmp) { + pr_info("Trying to register already registered routing " + "algorithm: %s\n", bat_algo->name); + goto out; + } + + INIT_HLIST_NODE(&bat_algo->list); + hlist_add_head_rcu(&bat_algo->list, &bat_algo_list); + ret = 0; + +out: + return ret; +} + +int bat_algo_select(struct bat_priv *bat_priv, char *name) +{ + struct bat_algo *bat_algo; + int ret = -1; + + bat_algo = bat_algo_get(name); + if (!bat_algo) + goto out; + + /* TODO: must lock here - can we highjack an existing spinlock ? */ + rcu_assign_pointer(bat_priv->bat_algo, bat_algo); + ret = 0; + +out: + return ret; +} + +int bat_algo_seq_print_text(struct seq_file *seq, void *offset) +{ + struct bat_algo *bat_algo; + struct hlist_node *node; + + seq_printf(seq, "Available routing algorithms:\n"); + + rcu_read_lock(); + hlist_for_each_entry_rcu(bat_algo, node, &bat_algo_list, list) { + seq_printf(seq, "%s\n", bat_algo->name); + } + rcu_read_unlock(); + + return 0; +} + module_init(batman_init); module_exit(batman_exit); diff --git a/main.h b/main.h index 464439f..1fa71d8 100644 --- a/main.h +++ b/main.h @@ -159,6 +159,11 @@ void mesh_free(struct net_device *soft_iface); void inc_module_count(void); void dec_module_count(void); int is_my_mac(const uint8_t *addr); +int bat_algo_register(struct bat_algo *bat_algo); +int bat_algo_select(struct bat_priv *bat_priv, char *name); +int bat_algo_seq_print_text(struct seq_file *seq, void *offset); + +int bat_iv_init(void); #ifdef CONFIG_BATMAN_ADV_DEBUG int debug_log(struct bat_priv *bat_priv, const char *fmt, ...) __printf(2, 3); diff --git a/soft-interface.c b/soft-interface.c index bd8c7cf..a0046ac 100644 --- a/soft-interface.c +++ b/soft-interface.c @@ -855,6 +855,10 @@ struct net_device *softif_create(const char *name) bat_priv->primary_if = NULL; bat_priv->num_ifaces = 0; + ret = bat_algo_select(bat_priv, "BATMAN IV"); + if (ret < 0) + goto unreg_soft_iface; + ret = sysfs_add_meshif(soft_iface); if (ret < 0) goto unreg_soft_iface; diff --git a/types.h b/types.h index 35085f4..cfb0cc2 100644 --- a/types.h +++ b/types.h @@ -206,6 +206,7 @@ struct bat_priv { atomic_t gw_reselect; struct hard_iface __rcu *primary_if; /* rcu protected pointer */ struct vis_info *my_vis_info; + struct bat_algo __rcu *bat_algo; }; struct socket_client { @@ -344,4 +345,10 @@ struct softif_neigh { struct rcu_head rcu; }; + +struct bat_algo { + struct hlist_node list; + char name[20]; +}; + #endif /* _NET_BATMAN_ADV_TYPES_H_ */ -- 1.7.5.4 ^ permalink raw reply related [flat|nested] 19+ messages in thread
* [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change " Marek Lindner @ 2011-11-28 16:28 ` Marek Lindner 2011-11-28 19:02 ` Andrew Lunn 2011-11-29 15:23 ` Simon Wunderlich 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 3/3] batman-adv: allowing changing the routing algorithm via sysfs Marek Lindner 2011-11-28 19:00 ` [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change routing algorithm at runtime Andrew Lunn 2 siblings, 2 replies; 19+ messages in thread From: Marek Lindner @ 2011-11-28 16:28 UTC (permalink / raw) To: b.a.t.m.a.n; +Cc: Marek Lindner Signed-off-by: Marek Lindner <lindner_marek@yahoo.de> --- bat_iv_ogm.c | 21 +++++++++++++-------- bat_ogm.h | 35 ----------------------------------- hard-interface.c | 16 ++++++++++------ routing.c | 7 +++++-- send.c | 8 +++++--- types.h | 10 +++++++++- 6 files changed, 42 insertions(+), 55 deletions(-) delete mode 100644 bat_ogm.h diff --git a/bat_iv_ogm.c b/bat_iv_ogm.c index a2b25ad..8e4c81f 100644 --- a/bat_iv_ogm.c +++ b/bat_iv_ogm.c @@ -20,7 +20,6 @@ */ #include "main.h" -#include "bat_ogm.h" #include "translation-table.h" #include "ring_buffer.h" #include "originator.h" @@ -30,7 +29,7 @@ #include "hard-interface.h" #include "send.h" -void bat_ogm_init(struct hard_iface *hard_iface) +static void bat_ogm_init(struct hard_iface *hard_iface) { struct batman_ogm_packet *batman_ogm_packet; @@ -47,7 +46,7 @@ void bat_ogm_init(struct hard_iface *hard_iface) batman_ogm_packet->ttvn = 0; } -void bat_ogm_init_primary(struct hard_iface *hard_iface) +static void bat_ogm_init_primary(struct hard_iface *hard_iface) { struct batman_ogm_packet *batman_ogm_packet; @@ -56,7 +55,7 @@ void bat_ogm_init_primary(struct hard_iface *hard_iface) batman_ogm_packet->header.ttl = TTL; } -void bat_ogm_update_mac(struct hard_iface *hard_iface) +static void bat_ogm_update_mac(struct hard_iface *hard_iface) { struct batman_ogm_packet *batman_ogm_packet; @@ -157,7 +156,7 @@ static void bat_ogm_send_to_if(struct forw_packet *forw_packet, } /* send a batman ogm packet */ -void bat_ogm_emit(struct forw_packet *forw_packet) +static void bat_ogm_emit(struct forw_packet *forw_packet) { struct hard_iface *hard_iface; struct net_device *soft_iface; @@ -528,7 +527,7 @@ static void bat_ogm_forward(struct orig_node *orig_node, if_incoming, 0, bat_ogm_fwd_send_time()); } -void bat_ogm_schedule(struct hard_iface *hard_iface, int tt_num_changes) +static void bat_ogm_schedule(struct hard_iface *hard_iface, int tt_num_changes) { struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface); struct batman_ogm_packet *batman_ogm_packet; @@ -1140,7 +1139,7 @@ out: orig_node_free_ref(orig_node); } -void bat_ogm_receive(const struct ethhdr *ethhdr, unsigned char *packet_buff, +static void bat_ogm_receive(const struct ethhdr *ethhdr, unsigned char *packet_buff, int packet_len, struct hard_iface *if_incoming) { struct batman_ogm_packet *batman_ogm_packet; @@ -1170,8 +1169,14 @@ void bat_ogm_receive(const struct ethhdr *ethhdr, unsigned char *packet_buff, batman_ogm_packet->tt_num_changes)); } -static struct bat_algo bat_algo_iv = { +static struct bat_algo bat_algo_iv __read_mostly = { .name = "BATMAN IV", + .bat_ogm_init = bat_ogm_init, + .bat_ogm_init_primary = bat_ogm_init_primary, + .bat_ogm_update_mac = bat_ogm_update_mac, + .bat_ogm_schedule = bat_ogm_schedule, + .bat_ogm_emit = bat_ogm_emit, + .bat_ogm_receive = bat_ogm_receive, }; int __init bat_iv_init(void) diff --git a/bat_ogm.h b/bat_ogm.h deleted file mode 100644 index 69329c1..0000000 --- a/bat_ogm.h +++ /dev/null @@ -1,35 +0,0 @@ -/* - * Copyright (C) 2007-2011 B.A.T.M.A.N. contributors: - * - * Marek Lindner, Simon Wunderlich - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of version 2 of the GNU General Public - * License as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA - * 02110-1301, USA - * - */ - -#ifndef _NET_BATMAN_ADV_OGM_H_ -#define _NET_BATMAN_ADV_OGM_H_ - -#include "main.h" - -void bat_ogm_init(struct hard_iface *hard_iface); -void bat_ogm_init_primary(struct hard_iface *hard_iface); -void bat_ogm_update_mac(struct hard_iface *hard_iface); -void bat_ogm_schedule(struct hard_iface *hard_iface, int tt_num_changes); -void bat_ogm_emit(struct forw_packet *forw_packet); -void bat_ogm_receive(const struct ethhdr *ethhdr, unsigned char *packet_buff, - int packet_len, struct hard_iface *if_incoming); - -#endif /* _NET_BATMAN_ADV_OGM_H_ */ diff --git a/hard-interface.c b/hard-interface.c index d3e0e32..f4bc753 100644 --- a/hard-interface.c +++ b/hard-interface.c @@ -28,7 +28,6 @@ #include "bat_sysfs.h" #include "originator.h" #include "hash.h" -#include "bat_ogm.h" #include <linux/if_arp.h> @@ -147,7 +146,8 @@ static void primary_if_select(struct bat_priv *bat_priv, if (!new_hard_iface) return; - bat_ogm_init_primary(new_hard_iface); + if (bat_priv->bat_algo->bat_ogm_init_primary) + bat_priv->bat_algo->bat_ogm_init_primary(new_hard_iface); primary_if_update_addr(bat_priv); } @@ -233,7 +233,8 @@ static void hardif_activate_interface(struct hard_iface *hard_iface) bat_priv = netdev_priv(hard_iface->soft_iface); - bat_ogm_update_mac(hard_iface); + if (bat_priv->bat_algo->bat_ogm_update_mac) + bat_priv->bat_algo->bat_ogm_update_mac(hard_iface); hard_iface->if_status = IF_TO_BE_ACTIVATED; /** @@ -307,7 +308,8 @@ int hardif_enable_interface(struct hard_iface *hard_iface, hard_iface->soft_iface = soft_iface; bat_priv = netdev_priv(hard_iface->soft_iface); - bat_ogm_init(hard_iface); + if (bat_priv->bat_algo->bat_ogm_init) + bat_priv->bat_algo->bat_ogm_init(hard_iface); if (!hard_iface->packet_buff) { bat_err(hard_iface->soft_iface, "Can't add interface packet " @@ -527,9 +529,11 @@ static int hard_if_event(struct notifier_block *this, goto hardif_put; check_known_mac_addr(hard_iface->net_dev); - bat_ogm_update_mac(hard_iface); - bat_priv = netdev_priv(hard_iface->soft_iface); + + if (bat_priv->bat_algo->bat_ogm_update_mac) + bat_priv->bat_algo->bat_ogm_update_mac(hard_iface); + primary_if = primary_if_get_selected(bat_priv); if (!primary_if) goto hardif_put; diff --git a/routing.c b/routing.c index d5ddbd1..478b121 100644 --- a/routing.c +++ b/routing.c @@ -29,7 +29,6 @@ #include "originator.h" #include "vis.h" #include "unicast.h" -#include "bat_ogm.h" void slide_own_bcast_window(struct hard_iface *hard_iface) { @@ -248,6 +247,7 @@ int window_protected(struct bat_priv *bat_priv, int32_t seq_num_diff, int recv_bat_ogm_packet(struct sk_buff *skb, struct hard_iface *hard_iface) { + struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface); struct ethhdr *ethhdr; /* drop packet if it has not necessary minimum size */ @@ -274,7 +274,10 @@ int recv_bat_ogm_packet(struct sk_buff *skb, struct hard_iface *hard_iface) ethhdr = (struct ethhdr *)skb_mac_header(skb); - bat_ogm_receive(ethhdr, skb->data, skb_headlen(skb), hard_iface); + if (bat_priv->bat_algo->bat_ogm_receive) + bat_priv->bat_algo->bat_ogm_receive(ethhdr, skb->data, + skb_headlen(skb), + hard_iface); kfree_skb(skb); return NET_RX_SUCCESS; diff --git a/send.c b/send.c index b00a0f5..a8a7fc9 100644 --- a/send.c +++ b/send.c @@ -28,7 +28,6 @@ #include "vis.h" #include "gateway_common.h" #include "originator.h" -#include "bat_ogm.h" static void send_outstanding_bcast_packet(struct work_struct *work); @@ -168,7 +167,9 @@ void schedule_bat_ogm(struct hard_iface *hard_iface) if (primary_if) hardif_free_ref(primary_if); - bat_ogm_schedule(hard_iface, tt_num_changes); + if (bat_priv->bat_algo->bat_ogm_schedule) + bat_priv->bat_algo->bat_ogm_schedule(hard_iface, + tt_num_changes); } static void forw_packet_free(struct forw_packet *forw_packet) @@ -318,7 +319,8 @@ void send_outstanding_bat_ogm_packet(struct work_struct *work) if (atomic_read(&bat_priv->mesh_state) == MESH_DEACTIVATING) goto out; - bat_ogm_emit(forw_packet); + if (bat_priv->bat_algo->bat_ogm_emit) + bat_priv->bat_algo->bat_ogm_emit(forw_packet); /** * we have to have at least one packet in the queue diff --git a/types.h b/types.h index cfb0cc2..e3bdb85 100644 --- a/types.h +++ b/types.h @@ -345,10 +345,18 @@ struct softif_neigh { struct rcu_head rcu; }; - struct bat_algo { struct hlist_node list; char name[20]; + void (*bat_ogm_init)(struct hard_iface *hard_iface); + void (*bat_ogm_init_primary)(struct hard_iface *hard_iface); + void (*bat_ogm_update_mac)(struct hard_iface *hard_iface); + void (*bat_ogm_schedule)(struct hard_iface *hard_iface, + int tt_num_changes); + void (*bat_ogm_emit)(struct forw_packet *forw_packet); + void (*bat_ogm_receive)(const struct ethhdr *ethhdr, + unsigned char *packet_buff, int packet_len, + struct hard_iface *if_incoming); }; #endif /* _NET_BATMAN_ADV_TYPES_H_ */ -- 1.7.5.4 ^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure Marek Lindner @ 2011-11-28 19:02 ` Andrew Lunn 2011-11-28 19:45 ` Marek Lindner 2011-11-29 15:23 ` Simon Wunderlich 1 sibling, 1 reply; 19+ messages in thread From: Andrew Lunn @ 2011-11-28 19:02 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking; +Cc: Marek Lindner > +static struct bat_algo bat_algo_iv __read_mostly = { > .name = "BATMAN IV", > + .bat_ogm_init = bat_ogm_init, > + .bat_ogm_init_primary = bat_ogm_init_primary, > + .bat_ogm_update_mac = bat_ogm_update_mac, > + .bat_ogm_schedule = bat_ogm_schedule, > + .bat_ogm_emit = bat_ogm_emit, > + .bat_ogm_receive = bat_ogm_receive, > }; This needs an .owner = THIS_MODULE, so you can do reference counting on the module. Andrew ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure 2011-11-28 19:02 ` Andrew Lunn @ 2011-11-28 19:45 ` Marek Lindner 2011-11-29 6:09 ` Andrew Lunn 0 siblings, 1 reply; 19+ messages in thread From: Marek Lindner @ 2011-11-28 19:45 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking On Tuesday, November 29, 2011 03:02:45 Andrew Lunn wrote: > > +static struct bat_algo bat_algo_iv __read_mostly = { > > > > .name = "BATMAN IV", > > > > + .bat_ogm_init = bat_ogm_init, > > + .bat_ogm_init_primary = bat_ogm_init_primary, > > + .bat_ogm_update_mac = bat_ogm_update_mac, > > + .bat_ogm_schedule = bat_ogm_schedule, > > + .bat_ogm_emit = bat_ogm_emit, > > + .bat_ogm_receive = bat_ogm_receive, > > > > }; > > This needs an .owner = THIS_MODULE, > > so you can do reference counting on the module. It is not a separate module .. Cheers, Marek ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure 2011-11-28 19:45 ` Marek Lindner @ 2011-11-29 6:09 ` Andrew Lunn 2011-11-29 6:35 ` Marek Lindner 0 siblings, 1 reply; 19+ messages in thread From: Andrew Lunn @ 2011-11-29 6:09 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking On Tue, Nov 29, 2011 at 03:45:29AM +0800, Marek Lindner wrote: > On Tuesday, November 29, 2011 03:02:45 Andrew Lunn wrote: > > > +static struct bat_algo bat_algo_iv __read_mostly = { > > > > > > .name = "BATMAN IV", > > > > > > + .bat_ogm_init = bat_ogm_init, > > > + .bat_ogm_init_primary = bat_ogm_init_primary, > > > + .bat_ogm_update_mac = bat_ogm_update_mac, > > > + .bat_ogm_schedule = bat_ogm_schedule, > > > + .bat_ogm_emit = bat_ogm_emit, > > > + .bat_ogm_receive = bat_ogm_receive, > > > > > > }; > > > > This needs an .owner = THIS_MODULE, > > > > so you can do reference counting on the module. > > It is not a separate module .. Which leads to the question, why is it not a separate module? Andrew ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure 2011-11-29 6:09 ` Andrew Lunn @ 2011-11-29 6:35 ` Marek Lindner 2011-11-29 8:45 ` Andrew Lunn 0 siblings, 1 reply; 19+ messages in thread From: Marek Lindner @ 2011-11-29 6:35 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking On Tuesday, November 29, 2011 14:09:24 Andrew Lunn wrote: > On Tue, Nov 29, 2011 at 03:45:29AM +0800, Marek Lindner wrote: > > On Tuesday, November 29, 2011 03:02:45 Andrew Lunn wrote: > > > > +static struct bat_algo bat_algo_iv __read_mostly = { > > > > > > > > .name = "BATMAN IV", > > > > > > > > + .bat_ogm_init = bat_ogm_init, > > > > + .bat_ogm_init_primary = bat_ogm_init_primary, > > > > + .bat_ogm_update_mac = bat_ogm_update_mac, > > > > + .bat_ogm_schedule = bat_ogm_schedule, > > > > + .bat_ogm_emit = bat_ogm_emit, > > > > + .bat_ogm_receive = bat_ogm_receive, > > > > > > > > }; > > > > > > This needs an .owner = THIS_MODULE, > > > > > > so you can do reference counting on the module. > > > > It is not a separate module .. > > Which leads to the question, why is it not a separate module? Why should it be ? Cheers, Marek ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure 2011-11-29 6:35 ` Marek Lindner @ 2011-11-29 8:45 ` Andrew Lunn 0 siblings, 0 replies; 19+ messages in thread From: Andrew Lunn @ 2011-11-29 8:45 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking On Tue, Nov 29, 2011 at 02:35:09PM +0800, Marek Lindner wrote: > On Tuesday, November 29, 2011 14:09:24 Andrew Lunn wrote: > > On Tue, Nov 29, 2011 at 03:45:29AM +0800, Marek Lindner wrote: > > > On Tuesday, November 29, 2011 03:02:45 Andrew Lunn wrote: > > > > > +static struct bat_algo bat_algo_iv __read_mostly = { > > > > > > > > > > .name = "BATMAN IV", > > > > > > > > > > + .bat_ogm_init = bat_ogm_init, > > > > > + .bat_ogm_init_primary = bat_ogm_init_primary, > > > > > + .bat_ogm_update_mac = bat_ogm_update_mac, > > > > > + .bat_ogm_schedule = bat_ogm_schedule, > > > > > + .bat_ogm_emit = bat_ogm_emit, > > > > > + .bat_ogm_receive = bat_ogm_receive, > > > > > > > > > > }; > > > > > > > > This needs an .owner = THIS_MODULE, > > > > > > > > so you can do reference counting on the module. > > > > > > It is not a separate module .. > > > > Which leads to the question, why is it not a separate module? > > Why should it be ? I can think of a couple of reasons. 1) You might get asked when you post the code upstream, why you have the ability to register algorithms, but have not taken it further to allow modules to load algorithms. That seems to be the normal case in kernel code, registration implies modules. You probably need an argument ready why you don't want it. 2) It helps shows you have the split between the routing algorithm and the rest of the code correct. If you need to add lots of EXPORT_SYMBOL_GPL() for functions which the routing algorithm needs to call then maybe the abstraction is wrong? Andrew ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure Marek Lindner 2011-11-28 19:02 ` Andrew Lunn @ 2011-11-29 15:23 ` Simon Wunderlich 2011-12-04 20:00 ` Marek Lindner 1 sibling, 1 reply; 19+ messages in thread From: Simon Wunderlich @ 2011-11-29 15:23 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking [-- Attachment #1: Type: text/plain, Size: 1764 bytes --] Hi Marek, interesting approach, thanks for the patches! I don't really think that we need modules at this point, but we can easily extend it later - for example, net/mac80211/rate.c has a very similar API and works with modules. A few comments to the API: Generally, we should have a few lines per function in types.h to describe what each function is doing - we don't need that for a RFC patch, but that would be useful for the final version. On Tue, Nov 29, 2011 at 12:28:36AM +0800, Marek Lindner wrote: > struct bat_algo { > struct hlist_node list; > char name[20]; you can also use a simple pointer here: char *name; (the name will most probably static in the module anyways) > + void (*bat_ogm_init)(struct hard_iface *hard_iface); > + void (*bat_ogm_init_primary)(struct hard_iface *hard_iface); > + void (*bat_ogm_update_mac)(struct hard_iface *hard_iface); > + void (*bat_ogm_schedule)(struct hard_iface *hard_iface, > + int tt_num_changes); can't we put tt_num_changes somewhere bat_priv? > + void (*bat_ogm_emit)(struct forw_packet *forw_packet); > + void (*bat_ogm_receive)(const struct ethhdr *ethhdr, > + unsigned char *packet_buff, int packet_len, > + struct hard_iface *if_incoming); you can just pass the SKB instead of ethhdr, packet_buff, and packet_len -> they are derived from the skb anyway. Then, you have can do the API with always a hard_iface as the first parameter. ;) I also agree with Andrew on defining (a subset of) mandatory functions which every algo has to provide. The register function could then check and decline algos which don't define mandatory functions, and we don't have to check availability in the real code later ... best regards, Simon [-- Attachment #2: Digital signature --] [-- Type: application/pgp-signature, Size: 198 bytes --] ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure 2011-11-29 15:23 ` Simon Wunderlich @ 2011-12-04 20:00 ` Marek Lindner 2011-12-06 19:20 ` Simon Wunderlich 0 siblings, 1 reply; 19+ messages in thread From: Marek Lindner @ 2011-12-04 20:00 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking Hi, > Generally, we should have a few lines per function in types.h to > describe what each function is doing - we don't need that for a RFC patch, > but that would be useful for the final version. ok, I will do that. > > + void (*bat_ogm_init)(struct hard_iface *hard_iface); > > + void (*bat_ogm_init_primary)(struct hard_iface *hard_iface); > > + void (*bat_ogm_update_mac)(struct hard_iface *hard_iface); > > + void (*bat_ogm_schedule)(struct hard_iface *hard_iface, > > + int tt_num_changes); > > can't we put tt_num_changes somewhere bat_priv? tt_num_changes holds the number of changed TT entries since the last OGM. Obviously that changes all the time. Are you suggesting we store this value in bat_priv instead of a local variable ? > > + void (*bat_ogm_emit)(struct forw_packet *forw_packet); > > + void (*bat_ogm_receive)(const struct ethhdr *ethhdr, > > + unsigned char *packet_buff, int packet_len, > > + struct hard_iface *if_incoming); > > you can just pass the SKB instead of ethhdr, packet_buff, and packet_len > -> they are derived from the skb anyway. > > Then, you have can do the API with always a hard_iface as the first > parameter. ;) I will send a separate patch for this API change in a minute as it is not really a part of the dynamic routing algo patchset. Thanks for all the feedback! Cheers, Marek ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure 2011-12-04 20:00 ` Marek Lindner @ 2011-12-06 19:20 ` Simon Wunderlich 0 siblings, 0 replies; 19+ messages in thread From: Simon Wunderlich @ 2011-12-06 19:20 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking [-- Attachment #1: Type: text/plain, Size: 1119 bytes --] On Mon, Dec 05, 2011 at 04:00:48AM +0800, Marek Lindner wrote: > > > + void (*bat_ogm_init)(struct hard_iface *hard_iface); > > > + void (*bat_ogm_init_primary)(struct hard_iface *hard_iface); > > > + void (*bat_ogm_update_mac)(struct hard_iface *hard_iface); > > > + void (*bat_ogm_schedule)(struct hard_iface *hard_iface, > > > + int tt_num_changes); > > > > can't we put tt_num_changes somewhere bat_priv? > > tt_num_changes holds the number of changed TT entries since the last OGM. > Obviously that changes all the time. Are you suggesting we store this value in > bat_priv instead of a local variable ? > Yup. I know that it is changed all the time, just as tt_buff and tt_len changes - it is updated when the packet is generated. There is not really a technical advantage (or disadvantage), it would just be more clean if information like this is stored in bat_priv IMHO - we fetch other stuff like the tt buffer or the ogm from structs like bat_priv or hard interfaces. And maybe we want to change something in this mechanism in the future. Just my personal opinion ;) Simon [-- Attachment #2: Digital signature --] [-- Type: application/pgp-signature, Size: 198 bytes --] ^ permalink raw reply [flat|nested] 19+ messages in thread
* [B.A.T.M.A.N.] [PATCH 3/3] batman-adv: allowing changing the routing algorithm via sysfs 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change " Marek Lindner 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure Marek Lindner @ 2011-11-28 16:28 ` Marek Lindner 2011-11-28 19:00 ` [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change routing algorithm at runtime Andrew Lunn 2 siblings, 0 replies; 19+ messages in thread From: Marek Lindner @ 2011-11-28 16:28 UTC (permalink / raw) To: b.a.t.m.a.n; +Cc: Marek Lindner Signed-off-by: Marek Lindner <lindner_marek@yahoo.de> --- bat_sysfs.c | 35 +++++++++++++++++++++++++++++++++++ 1 files changed, 35 insertions(+), 0 deletions(-) diff --git a/bat_sysfs.c b/bat_sysfs.c index c25492f..5375600 100644 --- a/bat_sysfs.c +++ b/bat_sysfs.c @@ -272,6 +272,39 @@ static ssize_t store_vis_mode(struct kobject *kobj, struct attribute *attr, return count; } +static ssize_t show_bat_algo(struct kobject *kobj, struct attribute *attr, + char *buff) +{ + struct bat_priv *bat_priv = kobj_to_batpriv(kobj); + return sprintf(buff, "%s\n", bat_priv->bat_algo->name); +} + +static ssize_t store_bat_algo(struct kobject *kobj, struct attribute *attr, + char *buff, size_t count) +{ + struct net_device *net_dev = kobj_to_netdev(kobj); + struct bat_priv *bat_priv = netdev_priv(net_dev); + int ret; + + if (buff[count - 1] == '\n') + buff[count - 1] = '\0'; + else + buff[count] = '\0'; + + if (strcmp(bat_priv->bat_algo->name, buff) == 0) + return count; + + ret = bat_algo_select(bat_priv, buff); + if (ret < 0) + bat_info(net_dev, + "Invalid parameter for 'routing algorithm' setting " + "received: %s\n", buff); + else + bat_info(net_dev, "Changing routing algorithm to: %s\n", buff); + + return count; +} + static void post_gw_deselect(struct net_device *net_dev) { struct bat_priv *bat_priv = netdev_priv(net_dev); @@ -382,6 +415,7 @@ BAT_ATTR_BOOL(bonding, S_IRUGO | S_IWUSR, NULL); BAT_ATTR_BOOL(fragmentation, S_IRUGO | S_IWUSR, update_min_mtu); BAT_ATTR_BOOL(ap_isolation, S_IRUGO | S_IWUSR, NULL); static BAT_ATTR(vis_mode, S_IRUGO | S_IWUSR, show_vis_mode, store_vis_mode); +static BAT_ATTR(routing_algo, S_IRUGO | S_IWUSR, show_bat_algo, store_bat_algo); static BAT_ATTR(gw_mode, S_IRUGO | S_IWUSR, show_gw_mode, store_gw_mode); BAT_ATTR_UINT(orig_interval, S_IRUGO | S_IWUSR, 2 * JITTER, INT_MAX, NULL); BAT_ATTR_UINT(hop_penalty, S_IRUGO | S_IWUSR, 0, TQ_MAX_VALUE, NULL); @@ -399,6 +433,7 @@ static struct bat_attribute *mesh_attrs[] = { &bat_attr_fragmentation, &bat_attr_ap_isolation, &bat_attr_vis_mode, + &bat_attr_routing_algo, &bat_attr_gw_mode, &bat_attr_orig_interval, &bat_attr_hop_penalty, -- 1.7.5.4 ^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change routing algorithm at runtime 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change " Marek Lindner 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure Marek Lindner 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 3/3] batman-adv: allowing changing the routing algorithm via sysfs Marek Lindner @ 2011-11-28 19:00 ` Andrew Lunn 2 siblings, 0 replies; 19+ messages in thread From: Andrew Lunn @ 2011-11-28 19:00 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking; +Cc: Marek Lindner Hi Marek > +static struct bat_algo *bat_algo_get(char *name) > +{ > + struct bat_algo *bat_algo = NULL, *bat_algo_tmp; > + struct hlist_node *node; > + > + rcu_read_lock(); > + hlist_for_each_entry_rcu(bat_algo_tmp, node, &bat_algo_list, list) { > + if (strcmp(bat_algo_tmp->name, name) != 0) > + continue; > + > + bat_algo = bat_algo_tmp; You should have a try_module_get() here, to stop the module implementing the algorithm from being unloaded. Andrew ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [RFC] batman-adv: switch routing algorithm at runtime 2011-11-28 16:27 [B.A.T.M.A.N.] [RFC] batman-adv: switch routing algorithm at runtime Marek Lindner 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change " Marek Lindner @ 2011-11-28 18:47 ` Andrew Lunn 2011-11-28 19:42 ` Marek Lindner 1 sibling, 1 reply; 19+ messages in thread From: Andrew Lunn @ 2011-11-28 18:47 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking Hi Marek > * How do we design the algorithm registration. At the moment batman_init() > calls bat_iv_init() and possibly other algorithms in the future but this is > less than ideal. I had hoped to find a mechanism which allows an init function > to be declared and called inside of the respective routing algorithm files. > These files can be compiled into the module or not. I would suggest looking at some other modular system. Generally the core offers a _register() and a _unregister() function. The module implementing an algorithm would call the _register function in its module_init function and the _unregister in module_exit function. It also makes sense to have a naming scheme for these modules. It then becomes possible to use hotplug to load algorithms dynamically, when userspace requests an algorithm which is not already loaded. I guess you first need an algorithm when the soft interface is taken up. If you don't have an algorithm loaded at that point, it makes sense to have a hard coded (or maybe menuconfig) default algorithm, which you try to load via hotplug. > * The API and its implementation. It is highly likely that this API won't > survive for very long as it has been designed to satisfy one routing algorithm > only. However, changing it in the future should be easy. It would be good to comment in the header file which functions can be NULL and which must be implemented. It would be best to assume that new methods added from now on will be NULL in older algorithms. There are a few options for handling this. Check the method is not NULL before calling it. Or the _register function can replace each NULL with a nop function and throw an BUG_ON() when a must have function is NULL. Andrew ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [RFC] batman-adv: switch routing algorithm at runtime 2011-11-28 18:47 ` [B.A.T.M.A.N.] [RFC] batman-adv: switch " Andrew Lunn @ 2011-11-28 19:42 ` Marek Lindner 2011-11-29 6:18 ` Andrew Lunn 0 siblings, 1 reply; 19+ messages in thread From: Marek Lindner @ 2011-11-28 19:42 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking Andrew, > I would suggest looking at some other modular system. Generally the > core offers a _register() and a _unregister() function. The module > implementing an algorithm would call the _register function in its > module_init function and the _unregister in module_exit function. It > also makes sense to have a naming scheme for these modules. It then > becomes possible to use hotplug to load algorithms dynamically, when > userspace requests an algorithm which is not already loaded. I guess > you first need an algorithm when the soft interface is taken up. If > you don't have an algorithm loaded at that point, it makes sense to > have a hard coded (or maybe menuconfig) default algorithm, which you > try to load via hotplug. not sure where we got out of alignment. If you check the patches you will see that it is a modular design. As I described in my initial mail, each routing algorithm is supposed to register itself. The functions / default algorithm / etc already there .. And no, I am not going to implement a hotplugging system. At first I wanted a compile time option before I was convinced to work on a system which allows to switch the routing algorithm at runtime. This is as far as I will go. Feel free to propose your own patches if want this feature. > > * The API and its implementation. It is highly likely that this API won't > > survive for very long as it has been designed to satisfy one routing > > algorithm only. However, changing it in the future should be easy. > > It would be good to comment in the header file which functions can be > NULL and which must be implemented. It would be best to assume that > new methods added from now on will be NULL in older algorithms. There > are a few options for handling this. Check the method is not NULL > before calling it. Or the _register function can replace each NULL > with a nop function and throw an BUG_ON() when a must have function is > NULL. All functions are checked right now and all functions can be NULL. Check the patch #2 ... Cheers, Marek ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [RFC] batman-adv: switch routing algorithm at runtime 2011-11-28 19:42 ` Marek Lindner @ 2011-11-29 6:18 ` Andrew Lunn 2011-11-29 6:34 ` Marek Lindner 0 siblings, 1 reply; 19+ messages in thread From: Andrew Lunn @ 2011-11-29 6:18 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking > All functions are checked right now and all functions can be NULL. > Check the patch #2 ... This seems strange. What sort of routing protocol can you build, which has a NULL receive function? Only static routes. I would perform the check once in _register(), that there is a receive function, and never check it again. Save yourself a few cycles. There is an LWN article about these ideas. Worth a read. I will post a link later. Andrew ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [RFC] batman-adv: switch routing algorithm at runtime 2011-11-29 6:18 ` Andrew Lunn @ 2011-11-29 6:34 ` Marek Lindner 2011-11-29 8:02 ` Andrew Lunn 0 siblings, 1 reply; 19+ messages in thread From: Marek Lindner @ 2011-11-29 6:34 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking On Tuesday, November 29, 2011 14:18:10 Andrew Lunn wrote: > > All functions are checked right now and all functions can be NULL. > > Check the patch #2 ... > > This seems strange. What sort of routing protocol can you build, which > has a NULL receive function? Only static routes. I would perform the > check once in _register(), that there is a receive function, and never > check it again. Save yourself a few cycles. > > There is an LWN article about these ideas. Worth a read. I will post a > link later. Thanks. That will be helpful! Cheers, Marek ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [B.A.T.M.A.N.] [RFC] batman-adv: switch routing algorithm at runtime 2011-11-29 6:34 ` Marek Lindner @ 2011-11-29 8:02 ` Andrew Lunn 0 siblings, 0 replies; 19+ messages in thread From: Andrew Lunn @ 2011-11-29 8:02 UTC (permalink / raw) To: The list for a Better Approach To Mobile Ad-hoc Networking > > There is an LWN article about these ideas. Worth a read. I will post a > > link later. Here is the link, to part 1 https://lwn.net/Articles/444910/ and part 2 https://lwn.net/Articles/446317/ Andrew ^ permalink raw reply [flat|nested] 19+ messages in thread
* [B.A.T.M.A.N.] switch routing algorithm at runtime @ 2011-12-05 9:56 Marek Lindner 2011-12-05 9:56 ` [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure Marek Lindner 0 siblings, 1 reply; 19+ messages in thread From: Marek Lindner @ 2011-12-05 9:56 UTC (permalink / raw) To: b.a.t.m.a.n Hi, I massaged the RFC patches a little longer to integrate all the feedback received so far. They are ready to be considered 'real' patches now. The 'dynamically load routing kernel module' feature has not found its way into my patchset. At this point I still see no benefit and only complications. The mentioned reasons (kernel folks might not like it / helps to better abstract) can be addressed in a different way (for example asking David if he is going to accept it or not). The only reason seems to be "because we can". That's not enough for me. This does not mean we can't add this feature at a later point in time when we have the feeling it actually brings some benefits. Here the changelog: * sysfs documentation added * changed batman iv function prefix to bat_iv * renamed struct bat_algo to struct bat_algo_ops * all bat_algo_ops callbacks are mandatory for now * added struct bat_algo_ops documentation * introduced bat_algo.h for the routing algo init calls * bat_algo_ops->name became a pointer * removed all (unnecessary) locking from bat_algo_* Cheers, Marek ^ permalink raw reply [flat|nested] 19+ messages in thread
* [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure 2011-12-05 9:56 [B.A.T.M.A.N.] " Marek Lindner @ 2011-12-05 9:56 ` Marek Lindner 0 siblings, 0 replies; 19+ messages in thread From: Marek Lindner @ 2011-12-05 9:56 UTC (permalink / raw) To: b.a.t.m.a.n; +Cc: Marek Lindner Signed-off-by: Marek Lindner <lindner_marek@yahoo.de> --- bat_iv_ogm.c | 179 ++++++++++++++++++++++++++++-------------------------- bat_ogm.h | 34 ---------- hard-interface.c | 10 ++-- main.c | 12 ++++ routing.c | 4 +- send.c | 5 +- types.h | 14 ++++ 7 files changed, 129 insertions(+), 129 deletions(-) delete mode 100644 bat_ogm.h diff --git a/bat_iv_ogm.c b/bat_iv_ogm.c index 1847efa..1c483a5 100644 --- a/bat_iv_ogm.c +++ b/bat_iv_ogm.c @@ -20,7 +20,6 @@ */ #include "main.h" -#include "bat_ogm.h" #include "translation-table.h" #include "ring_buffer.h" #include "originator.h" @@ -31,7 +30,7 @@ #include "send.h" #include "bat_algo.h" -void bat_ogm_init(struct hard_iface *hard_iface) +static void bat_iv_ogm_init(struct hard_iface *hard_iface) { struct batman_ogm_packet *batman_ogm_packet; @@ -48,7 +47,7 @@ void bat_ogm_init(struct hard_iface *hard_iface) batman_ogm_packet->ttvn = 0; } -void bat_ogm_init_primary(struct hard_iface *hard_iface) +static void bat_iv_ogm_init_primary(struct hard_iface *hard_iface) { struct batman_ogm_packet *batman_ogm_packet; @@ -57,7 +56,7 @@ void bat_ogm_init_primary(struct hard_iface *hard_iface) batman_ogm_packet->header.ttl = TTL; } -void bat_ogm_update_mac(struct hard_iface *hard_iface) +static void bat_iv_ogm_update_mac(struct hard_iface *hard_iface) { struct batman_ogm_packet *batman_ogm_packet; @@ -69,7 +68,7 @@ void bat_ogm_update_mac(struct hard_iface *hard_iface) } /* when do we schedule our own ogm to be sent */ -static unsigned long bat_ogm_emit_send_time(const struct bat_priv *bat_priv) +static unsigned long bat_iv_ogm_emit_send_time(const struct bat_priv *bat_priv) { return jiffies + msecs_to_jiffies( atomic_read(&bat_priv->orig_interval) - @@ -77,7 +76,7 @@ static unsigned long bat_ogm_emit_send_time(const struct bat_priv *bat_priv) } /* when do we schedule a ogm packet to be sent */ -static unsigned long bat_ogm_fwd_send_time(void) +static unsigned long bat_iv_ogm_fwd_send_time(void) { return jiffies + msecs_to_jiffies(random32() % (JITTER/2)); } @@ -90,8 +89,8 @@ static uint8_t hop_penalty(uint8_t tq, const struct bat_priv *bat_priv) } /* is there another aggregated packet here? */ -static int bat_ogm_aggr_packet(int buff_pos, int packet_len, - int tt_num_changes) +static int bat_iv_ogm_aggr_packet(int buff_pos, int packet_len, + int tt_num_changes) { int next_buff_pos = buff_pos + BATMAN_OGM_LEN + tt_len(tt_num_changes); @@ -100,8 +99,8 @@ static int bat_ogm_aggr_packet(int buff_pos, int packet_len, } /* send a batman ogm to a given interface */ -static void bat_ogm_send_to_if(struct forw_packet *forw_packet, - struct hard_iface *hard_iface) +static void bat_iv_ogm_send_to_if(struct forw_packet *forw_packet, + struct hard_iface *hard_iface) { struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface); char *fwd_str; @@ -118,8 +117,8 @@ static void bat_ogm_send_to_if(struct forw_packet *forw_packet, batman_ogm_packet = (struct batman_ogm_packet *)forw_packet->skb->data; /* adjust all flags and log packets */ - while (bat_ogm_aggr_packet(buff_pos, forw_packet->packet_len, - batman_ogm_packet->tt_num_changes)) { + while (bat_iv_ogm_aggr_packet(buff_pos, forw_packet->packet_len, + batman_ogm_packet->tt_num_changes)) { /* we might have aggregated direct link packets with an * ordinary base packet */ @@ -158,7 +157,7 @@ static void bat_ogm_send_to_if(struct forw_packet *forw_packet, } /* send a batman ogm packet */ -void bat_ogm_emit(struct forw_packet *forw_packet) +static void bat_iv_ogm_emit(struct forw_packet *forw_packet) { struct hard_iface *hard_iface; struct net_device *soft_iface; @@ -217,7 +216,7 @@ void bat_ogm_emit(struct forw_packet *forw_packet) if (hard_iface->soft_iface != soft_iface) continue; - bat_ogm_send_to_if(forw_packet, hard_iface); + bat_iv_ogm_send_to_if(forw_packet, hard_iface); } rcu_read_unlock(); @@ -227,13 +226,13 @@ out: } /* return true if new_packet can be aggregated with forw_packet */ -static bool bat_ogm_can_aggregate(const struct batman_ogm_packet +static bool bat_iv_ogm_can_aggregate(const struct batman_ogm_packet *new_batman_ogm_packet, - struct bat_priv *bat_priv, - int packet_len, unsigned long send_time, - bool directlink, - const struct hard_iface *if_incoming, - const struct forw_packet *forw_packet) + struct bat_priv *bat_priv, + int packet_len, unsigned long send_time, + bool directlink, + const struct hard_iface *if_incoming, + const struct forw_packet *forw_packet) { struct batman_ogm_packet *batman_ogm_packet; int aggregated_bytes = forw_packet->packet_len + packet_len; @@ -307,11 +306,11 @@ out: } /* create a new aggregated packet and add this packet to it */ -static void bat_ogm_aggregate_new(const unsigned char *packet_buff, - int packet_len, unsigned long send_time, - bool direct_link, - struct hard_iface *if_incoming, - int own_packet) +static void bat_iv_ogm_aggregate_new(const unsigned char *packet_buff, + int packet_len, unsigned long send_time, + bool direct_link, + struct hard_iface *if_incoming, + int own_packet) { struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface); struct forw_packet *forw_packet_aggr; @@ -386,9 +385,9 @@ out: } /* aggregate a new packet into the existing ogm packet */ -static void bat_ogm_aggregate(struct forw_packet *forw_packet_aggr, - const unsigned char *packet_buff, - int packet_len, bool direct_link) +static void bat_iv_ogm_aggregate(struct forw_packet *forw_packet_aggr, + const unsigned char *packet_buff, + int packet_len, bool direct_link) { unsigned char *skb_buff; @@ -403,10 +402,10 @@ static void bat_ogm_aggregate(struct forw_packet *forw_packet_aggr, (1 << forw_packet_aggr->num_packets); } -static void bat_ogm_queue_add(struct bat_priv *bat_priv, - unsigned char *packet_buff, - int packet_len, struct hard_iface *if_incoming, - int own_packet, unsigned long send_time) +static void bat_iv_ogm_queue_add(struct bat_priv *bat_priv, + unsigned char *packet_buff, + int packet_len, struct hard_iface *if_incoming, + int own_packet, unsigned long send_time) { /** * _aggr -> pointer to the packet we want to aggregate with @@ -426,11 +425,11 @@ static void bat_ogm_queue_add(struct bat_priv *bat_priv, if ((atomic_read(&bat_priv->aggregated_ogms)) && (!own_packet)) { hlist_for_each_entry(forw_packet_pos, tmp_node, &bat_priv->forw_bat_list, list) { - if (bat_ogm_can_aggregate(batman_ogm_packet, - bat_priv, packet_len, - send_time, direct_link, - if_incoming, - forw_packet_pos)) { + if (bat_iv_ogm_can_aggregate(batman_ogm_packet, + bat_priv, packet_len, + send_time, direct_link, + if_incoming, + forw_packet_pos)) { forw_packet_aggr = forw_packet_pos; break; } @@ -452,20 +451,20 @@ static void bat_ogm_queue_add(struct bat_priv *bat_priv, (atomic_read(&bat_priv->aggregated_ogms))) send_time += msecs_to_jiffies(MAX_AGGREGATION_MS); - bat_ogm_aggregate_new(packet_buff, packet_len, - send_time, direct_link, - if_incoming, own_packet); + bat_iv_ogm_aggregate_new(packet_buff, packet_len, + send_time, direct_link, + if_incoming, own_packet); } else { - bat_ogm_aggregate(forw_packet_aggr, packet_buff, packet_len, - direct_link); + bat_iv_ogm_aggregate(forw_packet_aggr, packet_buff, + packet_len, direct_link); spin_unlock_bh(&bat_priv->forw_bat_list_lock); } } -static void bat_ogm_forward(struct orig_node *orig_node, - const struct ethhdr *ethhdr, - struct batman_ogm_packet *batman_ogm_packet, - int directlink, struct hard_iface *if_incoming) +static void bat_iv_ogm_forward(struct orig_node *orig_node, + const struct ethhdr *ethhdr, + struct batman_ogm_packet *batman_ogm_packet, + int directlink, struct hard_iface *if_incoming) { struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface); struct neigh_node *router; @@ -524,12 +523,13 @@ static void bat_ogm_forward(struct orig_node *orig_node, else batman_ogm_packet->flags &= ~DIRECTLINK; - bat_ogm_queue_add(bat_priv, (unsigned char *)batman_ogm_packet, - BATMAN_OGM_LEN + tt_len(tt_num_changes), - if_incoming, 0, bat_ogm_fwd_send_time()); + bat_iv_ogm_queue_add(bat_priv, (unsigned char *)batman_ogm_packet, + BATMAN_OGM_LEN + tt_len(tt_num_changes), + if_incoming, 0, bat_iv_ogm_fwd_send_time()); } -void bat_ogm_schedule(struct hard_iface *hard_iface, int tt_num_changes) +static void bat_iv_ogm_schedule(struct hard_iface *hard_iface, + int tt_num_changes) { struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface); struct batman_ogm_packet *batman_ogm_packet; @@ -566,21 +566,22 @@ void bat_ogm_schedule(struct hard_iface *hard_iface, int tt_num_changes) atomic_inc(&hard_iface->seqno); slide_own_bcast_window(hard_iface); - bat_ogm_queue_add(bat_priv, hard_iface->packet_buff, - hard_iface->packet_len, hard_iface, 1, - bat_ogm_emit_send_time(bat_priv)); + bat_iv_ogm_queue_add(bat_priv, hard_iface->packet_buff, + hard_iface->packet_len, hard_iface, 1, + bat_iv_ogm_emit_send_time(bat_priv)); if (primary_if) hardif_free_ref(primary_if); } -static void bat_ogm_orig_update(struct bat_priv *bat_priv, - struct orig_node *orig_node, - const struct ethhdr *ethhdr, - const struct batman_ogm_packet +static void bat_iv_ogm_orig_update(struct bat_priv *bat_priv, + struct orig_node *orig_node, + const struct ethhdr *ethhdr, + const struct batman_ogm_packet *batman_ogm_packet, - struct hard_iface *if_incoming, - const unsigned char *tt_buff, int is_duplicate) + struct hard_iface *if_incoming, + const unsigned char *tt_buff, + int is_duplicate) { struct neigh_node *neigh_node = NULL, *tmp_neigh_node = NULL; struct neigh_node *router = NULL; @@ -715,10 +716,10 @@ out: neigh_node_free_ref(router); } -static int bat_ogm_calc_tq(struct orig_node *orig_node, - struct orig_node *orig_neigh_node, - struct batman_ogm_packet *batman_ogm_packet, - struct hard_iface *if_incoming) +static int bat_iv_ogm_calc_tq(struct orig_node *orig_node, + struct orig_node *orig_neigh_node, + struct batman_ogm_packet *batman_ogm_packet, + struct hard_iface *if_incoming) { struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface); struct neigh_node *neigh_node = NULL, *tmp_neigh_node; @@ -827,10 +828,10 @@ out: * -1 the packet is old and has been received while the seqno window * was protected. Caller should drop it. */ -static int bat_ogm_update_seqnos(const struct ethhdr *ethhdr, - const struct batman_ogm_packet +static int bat_iv_ogm_update_seqnos(const struct ethhdr *ethhdr, + const struct batman_ogm_packet *batman_ogm_packet, - const struct hard_iface *if_incoming) + const struct hard_iface *if_incoming) { struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface); struct orig_node *orig_node; @@ -892,10 +893,10 @@ out: return ret; } -static void bat_ogm_process(const struct ethhdr *ethhdr, - struct batman_ogm_packet *batman_ogm_packet, - const unsigned char *tt_buff, - struct hard_iface *if_incoming) +static void bat_iv_ogm_process(const struct ethhdr *ethhdr, + struct batman_ogm_packet *batman_ogm_packet, + const unsigned char *tt_buff, + struct hard_iface *if_incoming) { struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface); struct hard_iface *hard_iface; @@ -1033,8 +1034,8 @@ static void bat_ogm_process(const struct ethhdr *ethhdr, if (!orig_node) return; - is_duplicate = bat_ogm_update_seqnos(ethhdr, batman_ogm_packet, - if_incoming); + is_duplicate = bat_iv_ogm_update_seqnos(ethhdr, batman_ogm_packet, + if_incoming); if (is_duplicate == -1) { bat_dbg(DBG_BATMAN, bat_priv, @@ -1083,8 +1084,8 @@ static void bat_ogm_process(const struct ethhdr *ethhdr, goto out_neigh; } - is_bidirectional = bat_ogm_calc_tq(orig_node, orig_neigh_node, - batman_ogm_packet, if_incoming); + is_bidirectional = bat_iv_ogm_calc_tq(orig_node, orig_neigh_node, + batman_ogm_packet, if_incoming); bonding_save_primary(orig_node, orig_neigh_node, batman_ogm_packet); @@ -1094,16 +1095,16 @@ static void bat_ogm_process(const struct ethhdr *ethhdr, (!is_duplicate || ((orig_node->last_real_seqno == batman_ogm_packet->seqno) && (orig_node->last_ttl - 3 <= batman_ogm_packet->header.ttl)))) - bat_ogm_orig_update(bat_priv, orig_node, ethhdr, - batman_ogm_packet, if_incoming, - tt_buff, is_duplicate); + bat_iv_ogm_orig_update(bat_priv, orig_node, ethhdr, + batman_ogm_packet, if_incoming, + tt_buff, is_duplicate); /* is single hop (direct) neighbor */ if (is_single_hop_neigh) { /* mark direct link on incoming interface */ - bat_ogm_forward(orig_node, ethhdr, batman_ogm_packet, - 1, if_incoming); + bat_iv_ogm_forward(orig_node, ethhdr, batman_ogm_packet, + 1, if_incoming); bat_dbg(DBG_BATMAN, bat_priv, "Forwarding packet: " "rebroadcast neighbor packet with direct link flag\n"); @@ -1125,7 +1126,8 @@ static void bat_ogm_process(const struct ethhdr *ethhdr, bat_dbg(DBG_BATMAN, bat_priv, "Forwarding packet: rebroadcast originator packet\n"); - bat_ogm_forward(orig_node, ethhdr, batman_ogm_packet, 0, if_incoming); + bat_iv_ogm_forward(orig_node, ethhdr, batman_ogm_packet, + 0, if_incoming); out_neigh: if ((orig_neigh_node) && (!is_single_hop_neigh)) @@ -1141,7 +1143,8 @@ out: orig_node_free_ref(orig_node); } -void bat_ogm_receive(struct hard_iface *if_incoming, struct sk_buff *skb) +static void bat_iv_ogm_receive(struct hard_iface *if_incoming, + struct sk_buff *skb) { struct batman_ogm_packet *batman_ogm_packet; struct ethhdr *ethhdr; @@ -1162,20 +1165,26 @@ void bat_ogm_receive(struct hard_iface *if_incoming, struct sk_buff *skb) tt_buff = packet_buff + buff_pos + BATMAN_OGM_LEN; - bat_ogm_process(ethhdr, batman_ogm_packet, - tt_buff, if_incoming); + bat_iv_ogm_process(ethhdr, batman_ogm_packet, + tt_buff, if_incoming); buff_pos += BATMAN_OGM_LEN + tt_len(batman_ogm_packet->tt_num_changes); batman_ogm_packet = (struct batman_ogm_packet *) (packet_buff + buff_pos); - } while (bat_ogm_aggr_packet(buff_pos, packet_len, - batman_ogm_packet->tt_num_changes)); + } while (bat_iv_ogm_aggr_packet(buff_pos, packet_len, + batman_ogm_packet->tt_num_changes)); } static struct bat_algo_ops batman_iv __read_mostly = { .name = "BATMAN IV", + .bat_ogm_init = bat_iv_ogm_init, + .bat_ogm_init_primary = bat_iv_ogm_init_primary, + .bat_ogm_update_mac = bat_iv_ogm_update_mac, + .bat_ogm_schedule = bat_iv_ogm_schedule, + .bat_ogm_emit = bat_iv_ogm_emit, + .bat_ogm_receive = bat_iv_ogm_receive, }; int __init bat_iv_init(void) diff --git a/bat_ogm.h b/bat_ogm.h deleted file mode 100644 index 47edfde..0000000 --- a/bat_ogm.h +++ /dev/null @@ -1,34 +0,0 @@ -/* - * Copyright (C) 2007-2011 B.A.T.M.A.N. contributors: - * - * Marek Lindner, Simon Wunderlich - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of version 2 of the GNU General Public - * License as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA - * 02110-1301, USA - * - */ - -#ifndef _NET_BATMAN_ADV_OGM_H_ -#define _NET_BATMAN_ADV_OGM_H_ - -#include "main.h" - -void bat_ogm_init(struct hard_iface *hard_iface); -void bat_ogm_init_primary(struct hard_iface *hard_iface); -void bat_ogm_update_mac(struct hard_iface *hard_iface); -void bat_ogm_schedule(struct hard_iface *hard_iface, int tt_num_changes); -void bat_ogm_emit(struct forw_packet *forw_packet); -void bat_ogm_receive(struct hard_iface *if_incoming, struct sk_buff *skb); - -#endif /* _NET_BATMAN_ADV_OGM_H_ */ diff --git a/hard-interface.c b/hard-interface.c index d3e0e32..e66d8d6 100644 --- a/hard-interface.c +++ b/hard-interface.c @@ -28,7 +28,6 @@ #include "bat_sysfs.h" #include "originator.h" #include "hash.h" -#include "bat_ogm.h" #include <linux/if_arp.h> @@ -147,7 +146,7 @@ static void primary_if_select(struct bat_priv *bat_priv, if (!new_hard_iface) return; - bat_ogm_init_primary(new_hard_iface); + bat_priv->bat_algo_ops->bat_ogm_init_primary(new_hard_iface); primary_if_update_addr(bat_priv); } @@ -233,7 +232,7 @@ static void hardif_activate_interface(struct hard_iface *hard_iface) bat_priv = netdev_priv(hard_iface->soft_iface); - bat_ogm_update_mac(hard_iface); + bat_priv->bat_algo_ops->bat_ogm_update_mac(hard_iface); hard_iface->if_status = IF_TO_BE_ACTIVATED; /** @@ -307,7 +306,7 @@ int hardif_enable_interface(struct hard_iface *hard_iface, hard_iface->soft_iface = soft_iface; bat_priv = netdev_priv(hard_iface->soft_iface); - bat_ogm_init(hard_iface); + bat_priv->bat_algo_ops->bat_ogm_init(hard_iface); if (!hard_iface->packet_buff) { bat_err(hard_iface->soft_iface, "Can't add interface packet " @@ -527,9 +526,10 @@ static int hard_if_event(struct notifier_block *this, goto hardif_put; check_known_mac_addr(hard_iface->net_dev); - bat_ogm_update_mac(hard_iface); bat_priv = netdev_priv(hard_iface->soft_iface); + bat_priv->bat_algo_ops->bat_ogm_update_mac(hard_iface); + primary_if = primary_if_get_selected(bat_priv); if (!primary_if) goto hardif_put; diff --git a/main.c b/main.c index b554f63..b7bfa38 100644 --- a/main.c +++ b/main.c @@ -206,6 +206,18 @@ int bat_algo_register(struct bat_algo_ops *bat_algo_ops) goto out; } + /* all algorithms must implement all ops (for now) */ + if (!bat_algo_ops->bat_ogm_init || + !bat_algo_ops->bat_ogm_init_primary || + !bat_algo_ops->bat_ogm_update_mac || + !bat_algo_ops->bat_ogm_schedule || + !bat_algo_ops->bat_ogm_emit || + !bat_algo_ops->bat_ogm_receive) { + pr_info("Routing algo '%s' does not implement required ops\n", + bat_algo_ops->name); + goto out; + } + INIT_HLIST_NODE(&bat_algo_ops->list); hlist_add_head(&bat_algo_ops->list, &bat_algo_list); ret = 0; diff --git a/routing.c b/routing.c index 5bc41c8..b72d7f3 100644 --- a/routing.c +++ b/routing.c @@ -29,7 +29,6 @@ #include "originator.h" #include "vis.h" #include "unicast.h" -#include "bat_ogm.h" void slide_own_bcast_window(struct hard_iface *hard_iface) { @@ -248,6 +247,7 @@ int window_protected(struct bat_priv *bat_priv, int32_t seq_num_diff, int recv_bat_ogm_packet(struct sk_buff *skb, struct hard_iface *hard_iface) { + struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface); struct ethhdr *ethhdr; /* drop packet if it has not necessary minimum size */ @@ -272,7 +272,7 @@ int recv_bat_ogm_packet(struct sk_buff *skb, struct hard_iface *hard_iface) if (skb_linearize(skb) < 0) return NET_RX_DROP; - bat_ogm_receive(hard_iface, skb); + bat_priv->bat_algo_ops->bat_ogm_receive(hard_iface, skb); kfree_skb(skb); return NET_RX_SUCCESS; diff --git a/send.c b/send.c index b00a0f5..019337e 100644 --- a/send.c +++ b/send.c @@ -28,7 +28,6 @@ #include "vis.h" #include "gateway_common.h" #include "originator.h" -#include "bat_ogm.h" static void send_outstanding_bcast_packet(struct work_struct *work); @@ -168,7 +167,7 @@ void schedule_bat_ogm(struct hard_iface *hard_iface) if (primary_if) hardif_free_ref(primary_if); - bat_ogm_schedule(hard_iface, tt_num_changes); + bat_priv->bat_algo_ops->bat_ogm_schedule(hard_iface, tt_num_changes); } static void forw_packet_free(struct forw_packet *forw_packet) @@ -318,7 +317,7 @@ void send_outstanding_bat_ogm_packet(struct work_struct *work) if (atomic_read(&bat_priv->mesh_state) == MESH_DEACTIVATING) goto out; - bat_ogm_emit(forw_packet); + bat_priv->bat_algo_ops->bat_ogm_emit(forw_packet); /** * we have to have at least one packet in the queue diff --git a/types.h b/types.h index d21ff2c..650ce5f 100644 --- a/types.h +++ b/types.h @@ -348,6 +348,20 @@ struct softif_neigh { struct bat_algo_ops { struct hlist_node list; char *name; + /* init OGM when hard-interface is enabled */ + void (*bat_ogm_init)(struct hard_iface *hard_iface); + /* init primary OGM when primary interface is selected */ + void (*bat_ogm_init_primary)(struct hard_iface *hard_iface); + /* init mac addresses of the OGM belonging to this hard-interface */ + void (*bat_ogm_update_mac)(struct hard_iface *hard_iface); + /* prepare a new outgoing OGM for the send queue */ + void (*bat_ogm_schedule)(struct hard_iface *hard_iface, + int tt_num_changes); + /* send scheduled OGM */ + void (*bat_ogm_emit)(struct forw_packet *forw_packet); + /* receive incoming OGM */ + void (*bat_ogm_receive)(struct hard_iface *if_incoming, + struct sk_buff *skb); }; #endif /* _NET_BATMAN_ADV_TYPES_H_ */ -- 1.7.5.4 ^ permalink raw reply related [flat|nested] 19+ messages in thread
end of thread, other threads:[~2011-12-06 19:20 UTC | newest] Thread overview: 19+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2011-11-28 16:27 [B.A.T.M.A.N.] [RFC] batman-adv: switch routing algorithm at runtime Marek Lindner 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change " Marek Lindner 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure Marek Lindner 2011-11-28 19:02 ` Andrew Lunn 2011-11-28 19:45 ` Marek Lindner 2011-11-29 6:09 ` Andrew Lunn 2011-11-29 6:35 ` Marek Lindner 2011-11-29 8:45 ` Andrew Lunn 2011-11-29 15:23 ` Simon Wunderlich 2011-12-04 20:00 ` Marek Lindner 2011-12-06 19:20 ` Simon Wunderlich 2011-11-28 16:28 ` [B.A.T.M.A.N.] [PATCH 3/3] batman-adv: allowing changing the routing algorithm via sysfs Marek Lindner 2011-11-28 19:00 ` [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: add infrastructure to change routing algorithm at runtime Andrew Lunn 2011-11-28 18:47 ` [B.A.T.M.A.N.] [RFC] batman-adv: switch " Andrew Lunn 2011-11-28 19:42 ` Marek Lindner 2011-11-29 6:18 ` Andrew Lunn 2011-11-29 6:34 ` Marek Lindner 2011-11-29 8:02 ` Andrew Lunn -- strict thread matches above, loose matches on Subject: below -- 2011-12-05 9:56 [B.A.T.M.A.N.] " Marek Lindner 2011-12-05 9:56 ` [B.A.T.M.A.N.] [PATCH 2/3] batman-adv: convert batman iv algorithm to use dynamic infrastructure Marek Lindner
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox