bond_main.c 142 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
/*
 * originally based on the dummy device.
 *
 * Copyright 1999, Thomas Davis, tadavis@lbl.gov.
 * Licensed under the GPL. Based on dummy.c, and eql.c devices.
 *
 * bonding.c: an Ethernet Bonding driver
 *
 * This is useful to talk to a Cisco EtherChannel compatible equipment:
 *	Cisco 5500
 *	Sun Trunking (Solaris)
 *	Alteon AceDirector Trunks
 *	Linux Bonding
 *	and probably many L2 switches ...
 *
 * How it works:
 *    ifconfig bond0 ipaddress netmask up
 *      will setup a network device, with an ip address.  No mac address
 *	will be assigned at this time.  The hw mac address will come from
 *	the first slave bonded to the channel.  All slaves will then use
 *	this hw mac address.
 *
 *    ifconfig bond0 down
 *         will release all slaves, marking them as down.
 *
 *    ifenslave bond0 eth0
 *	will attach eth0 to bond0 as a slave.  eth0 hw mac address will either
 *	a: be used as initial mac address
 *	b: if a hw mac address already is there, eth0's hw mac address
 *	   will then be set from bond0.
 *
 * v0.1 - first working version.
 * v0.2 - changed stats to be calculated by summing slaves stats.
 *
 * Changes:
 * Arnaldo Carvalho de Melo <acme@conectiva.com.br>
 * - fix leaks on failure at bond_init
 *
 * 2000/09/30 - Willy Tarreau <willy at meta-x.org>
 *     - added trivial code to release a slave device.
 *     - fixed security bug (CAP_NET_ADMIN not checked)
 *     - implemented MII link monitoring to disable dead links :
 *       All MII capable slaves are checked every <miimon> milliseconds
 *       (100 ms seems good). This value can be changed by passing it to
 *       insmod. A value of zero disables the monitoring (default).
 *     - fixed an infinite loop in bond_xmit_roundrobin() when there's no
 *       good slave.
 *     - made the code hopefully SMP safe
 *
 * 2000/10/03 - Willy Tarreau <willy at meta-x.org>
 *     - optimized slave lists based on relevant suggestions from Thomas Davis
 *     - implemented active-backup method to obtain HA with two switches:
 *       stay as long as possible on the same active interface, while we
 *       also monitor the backup one (MII link status) because we want to know
 *       if we are able to switch at any time. ( pass "mode=1" to insmod )
 *     - lots of stress testings because we need it to be more robust than the
 *       wires ! :->
 *
 * 2000/10/09 - Willy Tarreau <willy at meta-x.org>
 *     - added up and down delays after link state change.
 *     - optimized the slaves chaining so that when we run forward, we never
 *       repass through the bond itself, but we can find it by searching
 *       backwards. Renders the deletion more difficult, but accelerates the
 *       scan.
 *     - smarter enslaving and releasing.
 *     - finer and more robust SMP locking
 *
 * 2000/10/17 - Willy Tarreau <willy at meta-x.org>
 *     - fixed two potential SMP race conditions
 *
 * 2000/10/18 - Willy Tarreau <willy at meta-x.org>
 *     - small fixes to the monitoring FSM in case of zero delays
 * 2000/11/01 - Willy Tarreau <willy at meta-x.org>
 *     - fixed first slave not automatically used in trunk mode.
 * 2000/11/10 : spelling of "EtherChannel" corrected.
 * 2000/11/13 : fixed a race condition in case of concurrent accesses to ioctl().
 * 2000/12/16 : fixed improper usage of rtnl_exlock_nowait().
 *
 * 2001/1/3 - Chad N. Tindel <ctindel at ieee dot org>
 *     - The bonding driver now simulates MII status monitoring, just like
 *       a normal network device.  It will show that the link is down iff
 *       every slave in the bond shows that their links are down.  If at least
 *       one slave is up, the bond's MII status will appear as up.
 *
 * 2001/2/7 - Chad N. Tindel <ctindel at ieee dot org>
 *     - Applications can now query the bond from user space to get
 *       information which may be useful.  They do this by calling
 *       the BOND_INFO_QUERY ioctl.  Once the app knows how many slaves
 *       are in the bond, it can call the BOND_SLAVE_INFO_QUERY ioctl to
 *       get slave specific information (# link failures, etc).  See
 *       <linux/if_bonding.h> for more details.  The structs of interest
 *       are ifbond and ifslave.
 *
 * 2001/4/5 - Chad N. Tindel <ctindel at ieee dot org>
 *     - Ported to 2.4 Kernel
 *
 * 2001/5/2 - Jeffrey E. Mast <jeff at mastfamily dot com>
 *     - When a device is detached from a bond, the slave device is no longer
 *       left thinking that is has a master.
 *
 * 2001/5/16 - Jeffrey E. Mast <jeff at mastfamily dot com>
 *     - memset did not appropriately initialized the bond rw_locks. Used
 *       rwlock_init to initialize to unlocked state to prevent deadlock when
 *       first attempting a lock
 *     - Called SET_MODULE_OWNER for bond device
 *
 * 2001/5/17 - Tim Anderson <tsa at mvista.com>
 *     - 2 paths for releasing for slave release; 1 through ioctl
 *       and 2) through close. Both paths need to release the same way.
 *     - the free slave in bond release is changing slave status before
 *       the free. The netdev_set_master() is intended to change slave state
 *       so it should not be done as part of the release process.
 *     - Simple rule for slave state at release: only the active in A/B and
 *       only one in the trunked case.
 *
 * 2001/6/01 - Tim Anderson <tsa at mvista.com>
 *     - Now call dev_close when releasing a slave so it doesn't screw up
 *       out routing table.
 *
 * 2001/6/01 - Chad N. Tindel <ctindel at ieee dot org>
 *     - Added /proc support for getting bond and slave information.
 *       Information is in /proc/net/<bond device>/info.
 *     - Changed the locking when calling bond_close to prevent deadlock.
 *
 * 2001/8/05 - Janice Girouard <girouard at us.ibm.com>
 *     - correct problem where refcnt of slave is not incremented in bond_ioctl
 *       so the system hangs when halting.
 *     - correct locking problem when unable to malloc in bond_enslave.
 *     - adding bond_xmit_xor logic.
 *     - adding multiple bond device support.
 *
 * 2001/8/13 - Erik Habbinga <erik_habbinga at hp dot com>
 *     - correct locking problem with rtnl_exlock_nowait
 *
 * 2001/8/23 - Janice Girouard <girouard at us.ibm.com>
 *     - bzero initial dev_bonds, to correct oops
 *     - convert SIOCDEVPRIVATE to new MII ioctl calls
 *
 * 2001/9/13 - Takao Indoh <indou dot takao at jp dot fujitsu dot com>
 *     - Add the BOND_CHANGE_ACTIVE ioctl implementation
 *
 * 2001/9/14 - Mark Huth <mhuth at mvista dot com>
 *     - Change MII_LINK_READY to not check for end of auto-negotiation,
 *       but only for an up link.
 *
 * 2001/9/20 - Chad N. Tindel <ctindel at ieee dot org>
 *     - Add the device field to bonding_t.  Previously the net_device
 *       corresponding to a bond wasn't available from the bonding_t
 *       structure.
 *
 * 2001/9/25 - Janice Girouard <girouard at us.ibm.com>
 *     - add arp_monitor for active backup mode
 *
 * 2001/10/23 - Takao Indoh <indou dot takao at jp dot fujitsu dot com>
 *     - Various memory leak fixes
 *
 * 2001/11/5 - Mark Huth <mark dot huth at mvista dot com>
 *     - Don't take rtnl lock in bond_mii_monitor as it deadlocks under
 *       certain hotswap conditions.
 *       Note:  this same change may be required in bond_arp_monitor ???
 *     - Remove possibility of calling bond_sethwaddr with NULL slave_dev ptr
 *     - Handle hot swap ethernet interface deregistration events to remove
 *       kernel oops following hot swap of enslaved interface
 *
 * 2002/1/2 - Chad N. Tindel <ctindel at ieee dot org>
 *     - Restore original slave flags at release time.
 *
 * 2002/02/18 - Erik Habbinga <erik_habbinga at hp dot com>
 *     - bond_release(): calling kfree on our_slave after call to
 *       bond_restore_slave_flags, not before
 *     - bond_enslave(): saving slave flags into original_flags before
 *       call to netdev_set_master, so the IFF_SLAVE flag doesn't end
 *       up in original_flags
 *
 * 2002/04/05 - Mark Smith <mark.smith at comdev dot cc> and
 *              Steve Mead <steve.mead at comdev dot cc>
 *     - Port Gleb Natapov's multicast support patchs from 2.4.12
 *       to 2.4.18 adding support for multicast.
 *
 * 2002/06/10 - Tony Cureington <tony.cureington * hp_com>
 *     - corrected uninitialized pointer (ifr.ifr_data) in bond_check_dev_link;
 *       actually changed function to use MIIPHY, then MIIREG, and finally
 *       ETHTOOL to determine the link status
 *     - fixed bad ifr_data pointer assignments in bond_ioctl
 *     - corrected mode 1 being reported as active-backup in bond_get_info;
 *       also added text to distinguish type of load balancing (rr or xor)
 *     - change arp_ip_target module param from "1-12s" (array of 12 ptrs)
 *       to "s" (a single ptr)
 *
 * 2002/08/30 - Jay Vosburgh <fubar at us dot ibm dot com>
 *     - Removed acquisition of xmit_lock in set_multicast_list; caused
 *       deadlock on SMP (lock is held by caller).
 *     - Revamped SIOCGMIIPHY, SIOCGMIIREG portion of bond_check_dev_link().
 *
 * 2002/09/18 - Jay Vosburgh <fubar at us dot ibm dot com>
 *     - Fixed up bond_check_dev_link() (and callers): removed some magic
 *	 numbers, banished local MII_ defines, wrapped ioctl calls to
 *	 prevent EFAULT errors
 *
 * 2002/9/30 - Jay Vosburgh <fubar at us dot ibm dot com>
 *     - make sure the ip target matches the arp_target before saving the
 *	 hw address.
 *
 * 2002/9/30 - Dan Eisner <eisner at 2robots dot com>
 *     - make sure my_ip is set before taking down the link, since
 *	 not all switches respond if the source ip is not set.
 *
 * 2002/10/8 - Janice Girouard <girouard at us dot ibm dot com>
 *     - read in the local ip address when enslaving a device
 *     - add primary support
 *     - make sure 2*arp_interval has passed when a new device
 *       is brought on-line before taking it down.
 *
 * 2002/09/11 - Philippe De Muyter <phdm at macqel dot be>
 *     - Added bond_xmit_broadcast logic.
 *     - Added bond_mode() support function.
 *
 * 2002/10/26 - Laurent Deniel <laurent.deniel at free.fr>
 *     - allow to register multicast addresses only on active slave
 *       (useful in active-backup mode)
 *     - add multicast module parameter
 *     - fix deletion of multicast groups after unloading module
 *
 * 2002/11/06 - Kameshwara Rayaprolu <kameshwara.rao * wipro_com>
 *     - Changes to prevent panic from closing the device twice; if we close
 *       the device in bond_release, we must set the original_flags to down
 *       so it won't be closed again by the network layer.
 *
 * 2002/11/07 - Tony Cureington <tony.cureington * hp_com>
 *     - Fix arp_target_hw_addr memory leak
 *     - Created activebackup_arp_monitor function to handle arp monitoring
 *       in active backup mode - the bond_arp_monitor had several problems...
 *       such as allowing slaves to tx arps sequentially without any delay
 *       for a response
 *     - Renamed bond_arp_monitor to loadbalance_arp_monitor and re-wrote
 *       this function to just handle arp monitoring in load-balancing mode;
 *       it is a lot more compact now
 *     - Changes to ensure one and only one slave transmits in active-backup
 *       mode
 *     - Robustesize parameters; warn users about bad combinations of
 *       parameters; also if miimon is specified and a network driver does
 *       not support MII or ETHTOOL, inform the user of this
 *     - Changes to support link_failure_count when in arp monitoring mode
 *     - Fix up/down delay reported in /proc
 *     - Added version; log version; make version available from "modinfo -d"
 *     - Fixed problem in bond_check_dev_link - if the first IOCTL (SIOCGMIIPH)
 *	 failed, the ETHTOOL ioctl never got a chance
 *
 * 2002/11/16 - Laurent Deniel <laurent.deniel at free.fr>
 *     - fix multicast handling in activebackup_arp_monitor
 *     - remove one unnecessary and confusing curr_active_slave == slave test
 *	 in activebackup_arp_monitor
 *
 *  2002/11/17 - Laurent Deniel <laurent.deniel at free.fr>
 *     - fix bond_slave_info_query when slave_id = num_slaves
 *
 *  2002/11/19 - Janice Girouard <girouard at us dot ibm dot com>
 *     - correct ifr_data reference.  Update ifr_data reference
 *       to mii_ioctl_data struct values to avoid confusion.
 *
 *  2002/11/22 - Bert Barbe <bert.barbe at oracle dot com>
 *      - Add support for multiple arp_ip_target
 *
 *  2002/12/13 - Jay Vosburgh <fubar at us dot ibm dot com>
 *	- Changed to allow text strings for mode and multicast, e.g.,
 *	  insmod bonding mode=active-backup.  The numbers still work.
 *	  One change: an invalid choice will cause module load failure,
 *	  rather than the previous behavior of just picking one.
 *	- Minor cleanups; got rid of dup ctype stuff, atoi function
 *
 * 2003/02/07 - Jay Vosburgh <fubar at us dot ibm dot com>
 *	- Added use_carrier module parameter that causes miimon to
 *	  use netif_carrier_ok() test instead of MII/ETHTOOL ioctls.
 *	- Minor cleanups; consolidated ioctl calls to one function.
 *
 * 2003/02/07 - Tony Cureington <tony.cureington * hp_com>
 *	- Fix bond_mii_monitor() logic error that could result in
 *	  bonding round-robin mode ignoring links after failover/recovery
 *
 * 2003/03/17 - Jay Vosburgh <fubar at us dot ibm dot com>
 *	- kmalloc fix (GFP_KERNEL to GFP_ATOMIC) reported by
 *	  Shmulik dot Hen at intel.com.
 *	- Based on discussion on mailing list, changed use of
 *	  update_slave_cnt(), created wrapper functions for adding/removing
 *	  slaves, changed bond_xmit_xor() to check slave_cnt instead of
 *	  checking slave and slave->dev (which only worked by accident).
 *	- Misc code cleanup: get arp_send() prototype from header file,
 *	  add max_bonds to bonding.txt.
 *
 * 2003/03/18 - Tsippy Mendelson <tsippy.mendelson at intel dot com> and
 *		Shmulik Hen <shmulik.hen at intel dot com>
 *	- Make sure only bond_attach_slave() and bond_detach_slave() can
 *	  manipulate the slave list, including slave_cnt, even when in
 *	  bond_release_all().
 *	- Fixed hang in bond_release() with traffic running:
 *	  netdev_set_master() must not be called from within the bond lock.
 *
 * 2003/03/18 - Tsippy Mendelson <tsippy.mendelson at intel dot com> and
 *		Shmulik Hen <shmulik.hen at intel dot com>
 *	- Fixed hang in bond_enslave() with traffic running:
 *	  netdev_set_master() must not be called from within the bond lock.
 *
 * 2003/03/18 - Amir Noam <amir.noam at intel dot com>
 *	- Added support for getting slave's speed and duplex via ethtool.
 *	  Needed for 802.3ad and other future modes.
 *
 * 2003/03/18 - Tsippy Mendelson <tsippy.mendelson at intel dot com> and
 *		Shmulik Hen <shmulik.hen at intel dot com>
 *	- Enable support of modes that need to use the unique mac address of
 *	  each slave.
 *	  * bond_enslave(): Moved setting the slave's mac address, and
 *	    openning it, from the application to the driver. This breaks
 *	    backward comaptibility with old versions of ifenslave that open
 *	     the slave before enalsving it !!!.
 *	  * bond_release(): The driver also takes care of closing the slave
 *	    and restoring its original mac address.
 *	- Removed the code that restores all base driver's flags.
 *	  Flags are automatically restored once all undo stages are done
 *	  properly.
 *	- Block possibility of enslaving before the master is up. This
 *	  prevents putting the system in an unstable state.
 *
 * 2003/03/18 - Amir Noam <amir.noam at intel dot com>,
 *		Tsippy Mendelson <tsippy.mendelson at intel dot com> and
 *		Shmulik Hen <shmulik.hen at intel dot com>
 *	- Added support for IEEE 802.3ad Dynamic link aggregation mode.
 *
 * 2003/05/01 - Amir Noam <amir.noam at intel dot com>
 *	- Added ABI version control to restore compatibility between
 *	  new/old ifenslave and new/old bonding.
 *
 * 2003/05/01 - Shmulik Hen <shmulik.hen at intel dot com>
 *	- Fixed bug in bond_release_all(): save old value of curr_active_slave
 *	  before setting it to NULL.
 *	- Changed driver versioning scheme to include version number instead
 *	  of release date (that is already in another field). There are 3
 *	  fields X.Y.Z where:
 *		X - Major version - big behavior changes
 *		Y - Minor version - addition of features
 *		Z - Extra version - minor changes and bug fixes
 *	  The current version is 1.0.0 as a base line.
 *
 * 2003/05/01 - Tsippy Mendelson <tsippy.mendelson at intel dot com> and
 *		Amir Noam <amir.noam at intel dot com>
 *	- Added support for lacp_rate module param.
 *	- Code beautification and style changes (mainly in comments).
 *	  new version - 1.0.1
 *
 * 2003/05/01 - Shmulik Hen <shmulik.hen at intel dot com>
 *	- Based on discussion on mailing list, changed locking scheme
 *	  to use lock/unlock or lock_bh/unlock_bh appropriately instead
 *	  of lock_irqsave/unlock_irqrestore. The new scheme helps exposing
 *	  hidden bugs and solves system hangs that occurred due to the fact
 *	  that holding lock_irqsave doesn't prevent softirqs from running.
 *	  This also increases total throughput since interrupts are not
 *	  blocked on each transmitted packets or monitor timeout.
 *	  new version - 2.0.0
 *
 * 2003/05/01 - Shmulik Hen <shmulik.hen at intel dot com>
 *	- Added support for Transmit load balancing mode.
 *	- Concentrate all assignments of curr_active_slave to a single point
 *	  so specific modes can take actions when the primary adapter is
 *	  changed.
 *	- Take the updelay parameter into consideration during bond_enslave
 *	  since some adapters loose their link during setting the device.
 *	- Renamed bond_3ad_link_status_changed() to
 *	  bond_3ad_handle_link_change() for compatibility with TLB.
 *	  new version - 2.1.0
 *
 * 2003/05/01 - Tsippy Mendelson <tsippy.mendelson at intel dot com>
 *	- Added support for Adaptive load balancing mode which is
 *	  equivalent to Transmit load balancing + Receive load balancing.
 *	  new version - 2.2.0
 *
 * 2003/05/15 - Jay Vosburgh <fubar at us dot ibm dot com>
 *	- Applied fix to activebackup_arp_monitor posted to bonding-devel
 *	  by Tony Cureington <tony.cureington * hp_com>.  Fixes ARP
 *	  monitor endless failover bug.  Version to 2.2.10
 *
 * 2003/05/20 - Amir Noam <amir.noam at intel dot com>
 *	- Fixed bug in ABI version control - Don't commit to a specific
 *	  ABI version if receiving unsupported ioctl commands.
 *
 * 2003/05/22 - Jay Vosburgh <fubar at us dot ibm dot com>
 *	- Fix ifenslave -c causing bond to loose existing routes;
 *	  added bond_set_mac_address() that doesn't require the
 *	  bond to be down.
 *	- In conjunction with fix for ifenslave -c, in
 *	  bond_change_active(), changing to the already active slave
 *	  is no longer an error (it successfully does nothing).
 *
 * 2003/06/30 - Amir Noam <amir.noam at intel dot com>
 * 	- Fixed bond_change_active() for ALB/TLB modes.
 *	  Version to 2.2.14.
 *
 * 2003/07/29 - Amir Noam <amir.noam at intel dot com>
 * 	- Fixed ARP monitoring bug.
 *	  Version to 2.2.15.
 *
 * 2003/07/31 - Willy Tarreau <willy at ods dot org>
 * 	- Fixed kernel panic when using ARP monitoring without
 *	  setting bond's IP address.
 *	  Version to 2.2.16.
 *
 * 2003/08/06 - Amir Noam <amir.noam at intel dot com>
 * 	- Back port from 2.6: use alloc_netdev(); fix /proc handling;
 *	  made stats a part of bond struct so no need to allocate
 *	  and free it separately; use standard list operations instead
 *	  of pre-allocated array of bonds.
 *	  Version to 2.3.0.
 *
 * 2003/08/07 - Jay Vosburgh <fubar at us dot ibm dot com>,
 *	       Amir Noam <amir.noam at intel dot com> and
 *	       Shmulik Hen <shmulik.hen at intel dot com>
 *	- Propagating master's settings: Distinguish between modes that
 *	  use a primary slave from those that don't, and propagate settings
 *	  accordingly; Consolidate change_active opeartions and add
 *	  reselect_active and find_best opeartions; Decouple promiscuous
 *	  handling from the multicast mode setting; Add support for changing
 *	  HW address and MTU with proper unwind; Consolidate procfs code,
 *	  add CHANGENAME handler; Enhance netdev notification handling.
 *	  Version to 2.4.0.
 *
 * 2003/09/15 - Stephen Hemminger <shemminger at osdl dot org>,
 *	       Amir Noam <amir.noam at intel dot com>
 *	- Convert /proc to seq_file interface.
 *	  Change /proc/net/bondX/info to /proc/net/bonding/bondX.
 *	  Set version to 2.4.1.
 *
 * 2003/11/20 - Amir Noam <amir.noam at intel dot com>
 *	- Fix /proc creation/destruction.
 *
 * 2003/12/01 - Shmulik Hen <shmulik.hen at intel dot com>
 *	- Massive cleanup - Set version to 2.5.0
 *	  Code changes:
 *	  o Consolidate format of prints and debug prints.
 *	  o Remove bonding_t/slave_t typedefs and consolidate all casts.
 *	  o Remove dead code and unnecessary checks.
 *	  o Consolidate starting/stopping timers.
 *	  o Consolidate handling of primary module param throughout the code.
 *	  o Removed multicast module param support - all settings are done
 *	    according to mode.
 *	  o Slave list iteration - bond is no longer part of the list,
 *	    added cyclic list iteration macros.
 *	  o Consolidate error handling in all xmit functions.
 *	  Style changes:
 *	  o Consolidate function naming and declarations.
 *	  o Consolidate function params and local variables names.
 *	  o Consolidate return values.
 *	  o Consolidate curly braces.
 *	  o Consolidate conditionals format.
 *	  o Change struct member names and types.
 *	  o Chomp trailing spaces, remove empty lines, fix indentations.
 *	  o Re-organize code according to context.
 *
 * 2003/12/30 - Amir Noam <amir.noam at intel dot com>
 *	- Fixed: Cannot remove and re-enslave the original active slave.
 *	- Fixed: Releasing the original active slave causes mac address
 *		 duplication.
 *	- Add support for slaves that use ethtool_ops.
 *	  Set version to 2.5.3.
 *
 * 2004/01/05 - Amir Noam <amir.noam at intel dot com>
 *	- Save bonding parameters per bond instead of using the global values.
 *	  Set version to 2.5.4.
 *
 * 2004/01/14 - Shmulik Hen <shmulik.hen at intel dot com>
 *	- Enhance VLAN support:
 *	  * Add support for VLAN hardware acceleration capable slaves.
 *	  * Add capability to tag self generated packets in ALB/TLB modes.
 *	  Set version to 2.6.0.
 * 2004/10/29 - Mitch Williams <mitch.a.williams at intel dot com>
 *      - Fixed bug when unloading module while using 802.3ad.  If
 *        spinlock debugging is turned on, this causes a stack dump.
 *        Solution is to move call to dev_remove_pack outside of the
 *        spinlock.
 *        Set version to 2.6.1.
Jay Vosburgh's avatar
Jay Vosburgh committed
478
479
480
481
 * 2005/06/05 - Jay Vosburgh <fubar@us.ibm.com>
 * 	- Support for generating gratuitous ARPs in active-backup mode.
 * 	  Includes support for VLAN tagging all bonding-generated ARPs
 * 	  as needed.  Set version to 2.6.2.
482
483
484
485
486
487
488
489
 * 2005/06/08 - Jason Gabler <jygabler at lbl dot gov>
 *	- alternate hashing policy support for mode 2
 *	  * Added kernel parameter "xmit_hash_policy" to allow the selection
 *	    of different hashing policies for mode 2.  The original mode 2
 *	    policy is the default, now found in xmit_hash_policy_layer2().
 *	  * Added xmit_hash_policy_layer34()
 *	- Modified by Jay Vosburgh <fubar@us.ibm.com> to also support mode 4.
 *	  Set version to 2.6.3.
Linus Torvalds's avatar
Linus Torvalds committed
490
491
492
493
494
495
496
497
498
499
500
501
502
503
 */

//#define BONDING_DEBUG 1

#include <linux/config.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/sched.h>
#include <linux/types.h>
#include <linux/fcntl.h>
#include <linux/interrupt.h>
#include <linux/ptrace.h>
#include <linux/ioport.h>
#include <linux/in.h>
504
#include <net/ip.h>
Linus Torvalds's avatar
Linus Torvalds committed
505
#include <linux/ip.h>
506
507
#include <linux/tcp.h>
#include <linux/udp.h>
Linus Torvalds's avatar
Linus Torvalds committed
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/init.h>
#include <linux/timer.h>
#include <linux/socket.h>
#include <linux/ctype.h>
#include <linux/inet.h>
#include <linux/bitops.h>
#include <asm/system.h>
#include <asm/io.h>
#include <asm/dma.h>
#include <asm/uaccess.h>
#include <linux/errno.h>
#include <linux/netdevice.h>
#include <linux/inetdevice.h>
#include <linux/etherdevice.h>
#include <linux/skbuff.h>
#include <net/sock.h>
#include <linux/rtnetlink.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/smp.h>
#include <linux/if_ether.h>
#include <net/arp.h>
#include <linux/mii.h>
#include <linux/ethtool.h>
#include <linux/if_vlan.h>
#include <linux/if_bonding.h>
Jay Vosburgh's avatar
Jay Vosburgh committed
536
#include <net/route.h>
Linus Torvalds's avatar
Linus Torvalds committed
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
#include "bonding.h"
#include "bond_3ad.h"
#include "bond_alb.h"

/*---------------------------- Module parameters ----------------------------*/

/* monitor all links that often (in milliseconds). <=0 disables monitoring */
#define BOND_LINK_MON_INTERV	0
#define BOND_LINK_ARP_INTERV	0

static int max_bonds	= BOND_DEFAULT_MAX_BONDS;
static int miimon	= BOND_LINK_MON_INTERV;
static int updelay	= 0;
static int downdelay	= 0;
static int use_carrier	= 1;
static char *mode	= NULL;
static char *primary	= NULL;
static char *lacp_rate	= NULL;
555
static char *xmit_hash_policy = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
static int arp_interval = BOND_LINK_ARP_INTERV;
static char *arp_ip_target[BOND_MAX_ARP_TARGETS] = { NULL, };

module_param(max_bonds, int, 0);
MODULE_PARM_DESC(max_bonds, "Max number of bonded devices");
module_param(miimon, int, 0);
MODULE_PARM_DESC(miimon, "Link check interval in milliseconds");
module_param(updelay, int, 0);
MODULE_PARM_DESC(updelay, "Delay before considering link up, in milliseconds");
module_param(downdelay, int, 0);
MODULE_PARM_DESC(downdelay, "Delay before considering link down, in milliseconds");
module_param(use_carrier, int, 0);
MODULE_PARM_DESC(use_carrier, "Use netif_carrier_ok (vs MII ioctls) in miimon; 0 for off, 1 for on (default)");
module_param(mode, charp, 0);
MODULE_PARM_DESC(mode, "Mode of operation : 0 for round robin, 1 for active-backup, 2 for xor");
module_param(primary, charp, 0);
MODULE_PARM_DESC(primary, "Primary network device to use");
module_param(lacp_rate, charp, 0);
MODULE_PARM_DESC(lacp_rate, "LACPDU tx rate to request from 802.3ad partner (slow/fast)");
575
576
module_param(xmit_hash_policy, charp, 0);
MODULE_PARM_DESC(xmit_hash_policy, "XOR hashing method : 0 for layer 2 (default), 1 for layer 3+4");
Linus Torvalds's avatar
Linus Torvalds committed
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
module_param(arp_interval, int, 0);
MODULE_PARM_DESC(arp_interval, "arp interval in milliseconds");
module_param_array(arp_ip_target, charp, NULL, 0);
MODULE_PARM_DESC(arp_ip_target, "arp targets in n.n.n.n form");

/*----------------------------- Global variables ----------------------------*/

static const char *version =
	DRV_DESCRIPTION ": v" DRV_VERSION " (" DRV_RELDATE ")\n";

static LIST_HEAD(bond_dev_list);

#ifdef CONFIG_PROC_FS
static struct proc_dir_entry *bond_proc_dir = NULL;
#endif

static u32 arp_target[BOND_MAX_ARP_TARGETS] = { 0, } ;
static int arp_ip_count	= 0;
static int bond_mode	= BOND_MODE_ROUNDROBIN;
596
static int xmit_hashtype= BOND_XMIT_POLICY_LAYER2;
Linus Torvalds's avatar
Linus Torvalds committed
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
static int lacp_fast	= 0;
static int app_abi_ver	= 0;
static int orig_app_abi_ver = -1; /* This is used to save the first ABI version
				   * we receive from the application. Once set,
				   * it won't be changed, and the module will
				   * refuse to enslave/release interfaces if the
				   * command comes from an application using
				   * another ABI version.
				   */
struct bond_parm_tbl {
	char *modename;
	int mode;
};

static struct bond_parm_tbl bond_lacp_tbl[] = {
{	"slow",		AD_LACP_SLOW},
{	"fast",		AD_LACP_FAST},
{	NULL,		-1},
};

static struct bond_parm_tbl bond_mode_tbl[] = {
{	"balance-rr",		BOND_MODE_ROUNDROBIN},
{	"active-backup",	BOND_MODE_ACTIVEBACKUP},
{	"balance-xor",		BOND_MODE_XOR},
{	"broadcast",		BOND_MODE_BROADCAST},
{	"802.3ad",		BOND_MODE_8023AD},
{	"balance-tlb",		BOND_MODE_TLB},
{	"balance-alb",		BOND_MODE_ALB},
{	NULL,			-1},
};

628
629
630
631
632
633
static struct bond_parm_tbl xmit_hashtype_tbl[] = {
{	"layer2",		BOND_XMIT_POLICY_LAYER2},
{	"layer3+4",		BOND_XMIT_POLICY_LAYER34},
{	NULL,			-1},
};

Linus Torvalds's avatar
Linus Torvalds committed
634
635
/*-------------------------- Forward declarations ---------------------------*/

636
static inline void bond_set_mode_ops(struct bonding *bond, int mode);
Jay Vosburgh's avatar
Jay Vosburgh committed
637
static void bond_send_gratuitous_arp(struct bonding *bond);
Linus Torvalds's avatar
Linus Torvalds committed
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685

/*---------------------------- General routines -----------------------------*/

static const char *bond_mode_name(int mode)
{
	switch (mode) {
	case BOND_MODE_ROUNDROBIN :
		return "load balancing (round-robin)";
	case BOND_MODE_ACTIVEBACKUP :
		return "fault-tolerance (active-backup)";
	case BOND_MODE_XOR :
		return "load balancing (xor)";
	case BOND_MODE_BROADCAST :
		return "fault-tolerance (broadcast)";
	case BOND_MODE_8023AD:
		return "IEEE 802.3ad Dynamic link aggregation";
	case BOND_MODE_TLB:
		return "transmit load balancing";
	case BOND_MODE_ALB:
		return "adaptive load balancing";
	default:
		return "unknown";
	}
}

/*---------------------------------- VLAN -----------------------------------*/

/**
 * bond_add_vlan - add a new vlan id on bond
 * @bond: bond that got the notification
 * @vlan_id: the vlan id to add
 *
 * Returns -ENOMEM if allocation failed.
 */
static int bond_add_vlan(struct bonding *bond, unsigned short vlan_id)
{
	struct vlan_entry *vlan;

	dprintk("bond: %s, vlan id %d\n",
		(bond ? bond->dev->name: "None"), vlan_id);

	vlan = kmalloc(sizeof(struct vlan_entry), GFP_KERNEL);
	if (!vlan) {
		return -ENOMEM;
	}

	INIT_LIST_HEAD(&vlan->vlan_list);
	vlan->vlan_id = vlan_id;
Jay Vosburgh's avatar
Jay Vosburgh committed
686
	vlan->vlan_ip = 0;
Linus Torvalds's avatar
Linus Torvalds committed
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000

	write_lock_bh(&bond->lock);

	list_add_tail(&vlan->vlan_list, &bond->vlan_list);

	write_unlock_bh(&bond->lock);

	dprintk("added VLAN ID %d on bond %s\n", vlan_id, bond->dev->name);

	return 0;
}

/**
 * bond_del_vlan - delete a vlan id from bond
 * @bond: bond that got the notification
 * @vlan_id: the vlan id to delete
 *
 * returns -ENODEV if @vlan_id was not found in @bond.
 */
static int bond_del_vlan(struct bonding *bond, unsigned short vlan_id)
{
	struct vlan_entry *vlan, *next;
	int res = -ENODEV;

	dprintk("bond: %s, vlan id %d\n", bond->dev->name, vlan_id);

	write_lock_bh(&bond->lock);

	list_for_each_entry_safe(vlan, next, &bond->vlan_list, vlan_list) {
		if (vlan->vlan_id == vlan_id) {
			list_del(&vlan->vlan_list);

			if ((bond->params.mode == BOND_MODE_TLB) ||
			    (bond->params.mode == BOND_MODE_ALB)) {
				bond_alb_clear_vlan(bond, vlan_id);
			}

			dprintk("removed VLAN ID %d from bond %s\n", vlan_id,
				bond->dev->name);

			kfree(vlan);

			if (list_empty(&bond->vlan_list) &&
			    (bond->slave_cnt == 0)) {
				/* Last VLAN removed and no slaves, so
				 * restore block on adding VLANs. This will
				 * be removed once new slaves that are not
				 * VLAN challenged will be added.
				 */
				bond->dev->features |= NETIF_F_VLAN_CHALLENGED;
			}

			res = 0;
			goto out;
		}
	}

	dprintk("couldn't find VLAN ID %d in bond %s\n", vlan_id,
		bond->dev->name);

out:
	write_unlock_bh(&bond->lock);
	return res;
}

/**
 * bond_has_challenged_slaves
 * @bond: the bond we're working on
 *
 * Searches the slave list. Returns 1 if a vlan challenged slave
 * was found, 0 otherwise.
 *
 * Assumes bond->lock is held.
 */
static int bond_has_challenged_slaves(struct bonding *bond)
{
	struct slave *slave;
	int i;

	bond_for_each_slave(bond, slave, i) {
		if (slave->dev->features & NETIF_F_VLAN_CHALLENGED) {
			dprintk("found VLAN challenged slave - %s\n",
				slave->dev->name);
			return 1;
		}
	}

	dprintk("no VLAN challenged slaves found\n");
	return 0;
}

/**
 * bond_next_vlan - safely skip to the next item in the vlans list.
 * @bond: the bond we're working on
 * @curr: item we're advancing from
 *
 * Returns %NULL if list is empty, bond->next_vlan if @curr is %NULL,
 * or @curr->next otherwise (even if it is @curr itself again).
 * 
 * Caller must hold bond->lock
 */
struct vlan_entry *bond_next_vlan(struct bonding *bond, struct vlan_entry *curr)
{
	struct vlan_entry *next, *last;

	if (list_empty(&bond->vlan_list)) {
		return NULL;
	}

	if (!curr) {
		next = list_entry(bond->vlan_list.next,
				  struct vlan_entry, vlan_list);
	} else {
		last = list_entry(bond->vlan_list.prev,
				  struct vlan_entry, vlan_list);
		if (last == curr) {
			next = list_entry(bond->vlan_list.next,
					  struct vlan_entry, vlan_list);
		} else {
			next = list_entry(curr->vlan_list.next,
					  struct vlan_entry, vlan_list);
		}
	}

	return next;
}

/**
 * bond_dev_queue_xmit - Prepare skb for xmit.
 * 
 * @bond: bond device that got this skb for tx.
 * @skb: hw accel VLAN tagged skb to transmit
 * @slave_dev: slave that is supposed to xmit this skbuff
 * 
 * When the bond gets an skb to transmit that is
 * already hardware accelerated VLAN tagged, and it
 * needs to relay this skb to a slave that is not
 * hw accel capable, the skb needs to be "unaccelerated",
 * i.e. strip the hwaccel tag and re-insert it as part
 * of the payload.
 */
int bond_dev_queue_xmit(struct bonding *bond, struct sk_buff *skb, struct net_device *slave_dev)
{
	unsigned short vlan_id;

	if (!list_empty(&bond->vlan_list) &&
	    !(slave_dev->features & NETIF_F_HW_VLAN_TX) &&
	    vlan_get_tag(skb, &vlan_id) == 0) {
		skb->dev = slave_dev;
		skb = vlan_put_tag(skb, vlan_id);
		if (!skb) {
			/* vlan_put_tag() frees the skb in case of error,
			 * so return success here so the calling functions
			 * won't attempt to free is again.
			 */
			return 0;
		}
	} else {
		skb->dev = slave_dev;
	}

	skb->priority = 1;
	dev_queue_xmit(skb);

	return 0;
}

/*
 * In the following 3 functions, bond_vlan_rx_register(), bond_vlan_rx_add_vid
 * and bond_vlan_rx_kill_vid, We don't protect the slave list iteration with a
 * lock because:
 * a. This operation is performed in IOCTL context,
 * b. The operation is protected by the RTNL semaphore in the 8021q code,
 * c. Holding a lock with BH disabled while directly calling a base driver
 *    entry point is generally a BAD idea.
 * 
 * The design of synchronization/protection for this operation in the 8021q
 * module is good for one or more VLAN devices over a single physical device
 * and cannot be extended for a teaming solution like bonding, so there is a
 * potential race condition here where a net device from the vlan group might
 * be referenced (either by a base driver or the 8021q code) while it is being
 * removed from the system. However, it turns out we're not making matters
 * worse, and if it works for regular VLAN usage it will work here too.
*/

/**
 * bond_vlan_rx_register - Propagates registration to slaves
 * @bond_dev: bonding net device that got called
 * @grp: vlan group being registered
 */
static void bond_vlan_rx_register(struct net_device *bond_dev, struct vlan_group *grp)
{
	struct bonding *bond = bond_dev->priv;
	struct slave *slave;
	int i;

	bond->vlgrp = grp;

	bond_for_each_slave(bond, slave, i) {
		struct net_device *slave_dev = slave->dev;

		if ((slave_dev->features & NETIF_F_HW_VLAN_RX) &&
		    slave_dev->vlan_rx_register) {
			slave_dev->vlan_rx_register(slave_dev, grp);
		}
	}
}

/**
 * bond_vlan_rx_add_vid - Propagates adding an id to slaves
 * @bond_dev: bonding net device that got called
 * @vid: vlan id being added
 */
static void bond_vlan_rx_add_vid(struct net_device *bond_dev, uint16_t vid)
{
	struct bonding *bond = bond_dev->priv;
	struct slave *slave;
	int i, res;

	bond_for_each_slave(bond, slave, i) {
		struct net_device *slave_dev = slave->dev;

		if ((slave_dev->features & NETIF_F_HW_VLAN_FILTER) &&
		    slave_dev->vlan_rx_add_vid) {
			slave_dev->vlan_rx_add_vid(slave_dev, vid);
		}
	}

	res = bond_add_vlan(bond, vid);
	if (res) {
		printk(KERN_ERR DRV_NAME
		       ": %s: Failed to add vlan id %d\n",
		       bond_dev->name, vid);
	}
}

/**
 * bond_vlan_rx_kill_vid - Propagates deleting an id to slaves
 * @bond_dev: bonding net device that got called
 * @vid: vlan id being removed
 */
static void bond_vlan_rx_kill_vid(struct net_device *bond_dev, uint16_t vid)
{
	struct bonding *bond = bond_dev->priv;
	struct slave *slave;
	struct net_device *vlan_dev;
	int i, res;

	bond_for_each_slave(bond, slave, i) {
		struct net_device *slave_dev = slave->dev;

		if ((slave_dev->features & NETIF_F_HW_VLAN_FILTER) &&
		    slave_dev->vlan_rx_kill_vid) {
			/* Save and then restore vlan_dev in the grp array,
			 * since the slave's driver might clear it.
			 */
			vlan_dev = bond->vlgrp->vlan_devices[vid];
			slave_dev->vlan_rx_kill_vid(slave_dev, vid);
			bond->vlgrp->vlan_devices[vid] = vlan_dev;
		}
	}

	res = bond_del_vlan(bond, vid);
	if (res) {
		printk(KERN_ERR DRV_NAME
		       ": %s: Failed to remove vlan id %d\n",
		       bond_dev->name, vid);
	}
}

static void bond_add_vlans_on_slave(struct bonding *bond, struct net_device *slave_dev)
{
	struct vlan_entry *vlan;

	write_lock_bh(&bond->lock);

	if (list_empty(&bond->vlan_list)) {
		goto out;
	}

	if ((slave_dev->features & NETIF_F_HW_VLAN_RX) &&
	    slave_dev->vlan_rx_register) {
		slave_dev->vlan_rx_register(slave_dev, bond->vlgrp);
	}

	if (!(slave_dev->features & NETIF_F_HW_VLAN_FILTER) ||
	    !(slave_dev->vlan_rx_add_vid)) {
		goto out;
	}

	list_for_each_entry(vlan, &bond->vlan_list, vlan_list) {
		slave_dev->vlan_rx_add_vid(slave_dev, vlan->vlan_id);
	}

out:
	write_unlock_bh(&bond->lock);
}

static void bond_del_vlans_from_slave(struct bonding *bond, struct net_device *slave_dev)
{
	struct vlan_entry *vlan;
	struct net_device *vlan_dev;

	write_lock_bh(&bond->lock);

	if (list_empty(&bond->vlan_list)) {
		goto out;
	}

	if (!(slave_dev->features & NETIF_F_HW_VLAN_FILTER) ||
	    !(slave_dev->vlan_rx_kill_vid)) {
		goto unreg;
	}