WSL2-Linux-Kernel/include/net/bonding.h

725 строки
19 KiB
C
Исходник Обычный вид История

/*
* Bond several ethernet interfaces into a Cisco, running 'Etherchannel'.
*
* Portions are (c) Copyright 1995 Simon "Guru Aleph-Null" Janes
* NCM: Network and Communications Management, Inc.
*
* BUT, I'm the one who modified it for ethernet, so:
* (c) Copyright 1999, Thomas Davis, tadavis@lbl.gov
*
* This software may be used and distributed according to the terms
* of the GNU Public License, incorporated herein by reference.
*
*/
#ifndef _NET_BONDING_H
#define _NET_BONDING_H
#include <linux/timer.h>
#include <linux/proc_fs.h>
#include <linux/if_bonding.h>
#include <linux/cpumask.h>
#include <linux/in6.h>
#include <linux/netpoll.h>
bonding: remove entries for master_ip and vlan_ip and query devices instead The following patch aimed to resolve an issue where secondary, tertiary, etc. addresses added to bond interfaces could overwrite the bond->master_ip and vlan_ip values. commit 917fbdb32f37e9a93b00bb12ee83532982982df3 Author: Henrik Saavedra Persson <henrik.e.persson@ericsson.com> Date: Wed Nov 23 23:37:15 2011 +0000 bonding: only use primary address for ARP That patch was good because it prevented bonds using ARP monitoring from sending frames with an invalid source IP address. Unfortunately, it didn't always work as expected. When using an ioctl (like ifconfig does) to set the IP address and netmask, 2 separate ioctls are actually called to set the IP and netmask if the mask chosen doesn't match the standard mask for that class of address. The first ioctl did not have a mask that matched the one in the primary address and would still cause the device address to be overwritten. The second ioctl that was called to set the mask would then detect as secondary and ignored, but the damage was already done. This was not an issue when using an application that used netlink sockets as the setting of IP and netmask came down at once. The inconsistent behavior between those two interfaces was something that needed to be resolved. While I was thinking about how I wanted to resolve this, Ralf Zeidler came with a patch that resolved this on a RHEL kernel by keeping a full shadow of the entries in dev->ifa_list for the bonding device and vlan devices in the bonding driver. I didn't like the duplication of the list as I want to see the 'bonding' struct and code shrink rather than grow, but liked the general idea. As the Subject indicates this patch drops the master_ip and vlan_ip elements from the 'bonding' and 'vlan_entry' structs, respectively. This can be done because a device's address-list is now traversed to determine the optimal source IP address for ARP requests and for checks to see if the bonding device has a particular IP address. This code could have all be contained inside the bonding driver, but it made more sense to me to EXPORT and call inet_confirm_addr since it did exactly what was needed. I tested this and a backported patch and everything works as expected. Ralf also helped with verification of the backported patch. Thanks to Ralf for all his help on this. v2: Whitespace and organizational changes based on suggestions from Jay Vosburgh and Dave Miller. v3: Fixup incorrect usage of rcu_read_unlock based on Dave Miller's suggestion. Signed-off-by: Andy Gospodarek <andy@greyhouse.net> CC: Ralf Zeidler <ralf.zeidler@nsn.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-22 20:14:29 +04:00
#include <linux/inetdevice.h>
#include <linux/etherdevice.h>
reciprocal_divide: update/correction of the algorithm Jakub Zawadzki noticed that some divisions by reciprocal_divide() were not correct [1][2], which he could also show with BPF code after divisions are transformed into reciprocal_value() for runtime invariance which can be passed to reciprocal_divide() later on; reverse in BPF dump ended up with a different, off-by-one K in some situations. This has been fixed by Eric Dumazet in commit aee636c4809fa5 ("bpf: do not use reciprocal divide"). This follow-up patch improves reciprocal_value() and reciprocal_divide() to work in all cases by using Granlund and Montgomery method, so that also future use is safe and without any non-obvious side-effects. Known problems with the old implementation were that division by 1 always returned 0 and some off-by-ones when the dividend and divisor where very large. This seemed to not be problematic with its current users, as far as we can tell. Eric Dumazet checked for the slab usage, we cannot surely say so in the case of flex_array. Still, in order to fix that, we propose an extension from the original implementation from commit 6a2d7a955d8d resp. [3][4], by using the algorithm proposed in "Division by Invariant Integers Using Multiplication" [5], Torbjörn Granlund and Peter L. Montgomery, that is, pseudocode for q = n/d where q, n, d is in u32 universe: 1) Initialization: int l = ceil(log_2 d) uword m' = floor((1<<32)*((1<<l)-d)/d)+1 int sh_1 = min(l,1) int sh_2 = max(l-1,0) 2) For q = n/d, all uword: uword t = (n*m')>>32 q = (t+((n-t)>>sh_1))>>sh_2 The assembler implementation from Agner Fog [6] also helped a lot while implementing. We have tested the implementation on x86_64, ppc64, i686, s390x; on x86_64/haswell we're still half the latency compared to normal divide. Joint work with Daniel Borkmann. [1] http://www.wireshark.org/~darkjames/reciprocal-buggy.c [2] http://www.wireshark.org/~darkjames/set-and-dump-filter-k-bug.c [3] https://gmplib.org/~tege/division-paper.pdf [4] http://homepage.cs.uiowa.edu/~jones/bcd/divide.html [5] http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.1.2556 [6] http://www.agner.org/optimize/asmlib.zip Reported-by: Jakub Zawadzki <darkjames-ws@darkjames.pl> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Austin S Hemmelgarn <ahferroin7@gmail.com> Cc: linux-kernel@vger.kernel.org Cc: Jesse Gross <jesse@nicira.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: Matt Mackall <mpm@selenic.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: Christoph Lameter <cl@linux-foundation.org> Cc: Andy Gospodarek <andy@greyhouse.net> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Jakub Zawadzki <darkjames-ws@darkjames.pl> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-22 05:29:41 +04:00
#include <linux/reciprocal_div.h>
#include <linux/if_link.h>
reciprocal_divide: update/correction of the algorithm Jakub Zawadzki noticed that some divisions by reciprocal_divide() were not correct [1][2], which he could also show with BPF code after divisions are transformed into reciprocal_value() for runtime invariance which can be passed to reciprocal_divide() later on; reverse in BPF dump ended up with a different, off-by-one K in some situations. This has been fixed by Eric Dumazet in commit aee636c4809fa5 ("bpf: do not use reciprocal divide"). This follow-up patch improves reciprocal_value() and reciprocal_divide() to work in all cases by using Granlund and Montgomery method, so that also future use is safe and without any non-obvious side-effects. Known problems with the old implementation were that division by 1 always returned 0 and some off-by-ones when the dividend and divisor where very large. This seemed to not be problematic with its current users, as far as we can tell. Eric Dumazet checked for the slab usage, we cannot surely say so in the case of flex_array. Still, in order to fix that, we propose an extension from the original implementation from commit 6a2d7a955d8d resp. [3][4], by using the algorithm proposed in "Division by Invariant Integers Using Multiplication" [5], Torbjörn Granlund and Peter L. Montgomery, that is, pseudocode for q = n/d where q, n, d is in u32 universe: 1) Initialization: int l = ceil(log_2 d) uword m' = floor((1<<32)*((1<<l)-d)/d)+1 int sh_1 = min(l,1) int sh_2 = max(l-1,0) 2) For q = n/d, all uword: uword t = (n*m')>>32 q = (t+((n-t)>>sh_1))>>sh_2 The assembler implementation from Agner Fog [6] also helped a lot while implementing. We have tested the implementation on x86_64, ppc64, i686, s390x; on x86_64/haswell we're still half the latency compared to normal divide. Joint work with Daniel Borkmann. [1] http://www.wireshark.org/~darkjames/reciprocal-buggy.c [2] http://www.wireshark.org/~darkjames/set-and-dump-filter-k-bug.c [3] https://gmplib.org/~tege/division-paper.pdf [4] http://homepage.cs.uiowa.edu/~jones/bcd/divide.html [5] http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.1.2556 [6] http://www.agner.org/optimize/asmlib.zip Reported-by: Jakub Zawadzki <darkjames-ws@darkjames.pl> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Austin S Hemmelgarn <ahferroin7@gmail.com> Cc: linux-kernel@vger.kernel.org Cc: Jesse Gross <jesse@nicira.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: Matt Mackall <mpm@selenic.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: Christoph Lameter <cl@linux-foundation.org> Cc: Andy Gospodarek <andy@greyhouse.net> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Jakub Zawadzki <darkjames-ws@darkjames.pl> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-22 05:29:41 +04:00
#include <net/bond_3ad.h>
#include <net/bond_alb.h>
#include <net/bond_options.h>
#define BOND_MAX_ARP_TARGETS 16
#define BOND_DEFAULT_MIIMON 100
#ifndef __long_aligned
#define __long_aligned __attribute__((aligned((sizeof(long)))))
#endif
#define BOND_MODE(bond) ((bond)->params.mode)
/* slave list primitives */
#define bond_slave_list(bond) (&(bond)->dev->adj_list.lower)
#define bond_has_slaves(bond) !list_empty(bond_slave_list(bond))
/* IMPORTANT: bond_first/last_slave can return NULL in case of an empty list */
#define bond_first_slave(bond) \
(bond_has_slaves(bond) ? \
netdev_adjacent_get_private(bond_slave_list(bond)->next) : \
NULL)
#define bond_last_slave(bond) \
(bond_has_slaves(bond) ? \
netdev_adjacent_get_private(bond_slave_list(bond)->prev) : \
NULL)
/* Caller must have rcu_read_lock */
#define bond_first_slave_rcu(bond) \
netdev_lower_get_first_private_rcu(bond->dev)
#define bond_is_first_slave(bond, pos) (pos == bond_first_slave(bond))
#define bond_is_last_slave(bond, pos) (pos == bond_last_slave(bond))
/**
* bond_for_each_slave - iterate over all slaves
* @bond: the bond holding this list
* @pos: current slave
* @iter: list_head * iterator
*
* Caller must hold RTNL
*/
#define bond_for_each_slave(bond, pos, iter) \
netdev_for_each_lower_private((bond)->dev, pos, iter)
bonding: initial RCU conversion This patch does the initial bonding conversion to RCU. After it the following modes are protected by RCU alone: roundrobin, active-backup, broadcast and xor. Modes ALB/TLB and 3ad still acquire bond->lock for reading, and will be dealt with later. curr_active_slave needs to be dereferenced via rcu in the converted modes because the only thing protecting the slave after this patch is rcu_read_lock, so we need the proper barrier for weakly ordered archs and to make sure we don't have stale pointer. It's not tagged with __rcu yet because there's still work to be done to remove the curr_slave_lock, so sparse will complain when rcu_assign_pointer and rcu_dereference are used, but the alternative to use rcu_dereference_protected would've created much bigger code churn which is more difficult to test and review. That will be converted in time. 1. Active-backup mode 1.1 Perf recording while doing iperf -P 4 - old bonding: iperf spent 0.55% in bonding, system spent 0.29% CPU in bonding - new bonding: iperf spent 0.29% in bonding, system spent 0.15% CPU in bonding 1.2. Bandwidth measurements - old bonding: 16.1 gbps consistently - new bonding: 17.5 gbps consistently 2. Round-robin mode 2.1 Perf recording while doing iperf -P 4 - old bonding: iperf spent 0.51% in bonding, system spent 0.24% CPU in bonding - new bonding: iperf spent 0.16% in bonding, system spent 0.11% CPU in bonding 2.2 Bandwidth measurements - old bonding: 8 gbps (variable due to packet reorderings) - new bonding: 10 gbps (variable due to packet reorderings) Of course the latency has improved in all converted modes, and moreover while doing enslave/release (since it doesn't affect tx anymore). Also I've stress tested all modes doing enslave/release in a loop while transmitting traffic. Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-01 18:54:51 +04:00
/* Caller must have rcu_read_lock */
#define bond_for_each_slave_rcu(bond, pos, iter) \
netdev_for_each_lower_private_rcu((bond)->dev, pos, iter)
bonding: initial RCU conversion This patch does the initial bonding conversion to RCU. After it the following modes are protected by RCU alone: roundrobin, active-backup, broadcast and xor. Modes ALB/TLB and 3ad still acquire bond->lock for reading, and will be dealt with later. curr_active_slave needs to be dereferenced via rcu in the converted modes because the only thing protecting the slave after this patch is rcu_read_lock, so we need the proper barrier for weakly ordered archs and to make sure we don't have stale pointer. It's not tagged with __rcu yet because there's still work to be done to remove the curr_slave_lock, so sparse will complain when rcu_assign_pointer and rcu_dereference are used, but the alternative to use rcu_dereference_protected would've created much bigger code churn which is more difficult to test and review. That will be converted in time. 1. Active-backup mode 1.1 Perf recording while doing iperf -P 4 - old bonding: iperf spent 0.55% in bonding, system spent 0.29% CPU in bonding - new bonding: iperf spent 0.29% in bonding, system spent 0.15% CPU in bonding 1.2. Bandwidth measurements - old bonding: 16.1 gbps consistently - new bonding: 17.5 gbps consistently 2. Round-robin mode 2.1 Perf recording while doing iperf -P 4 - old bonding: iperf spent 0.51% in bonding, system spent 0.24% CPU in bonding - new bonding: iperf spent 0.16% in bonding, system spent 0.11% CPU in bonding 2.2 Bandwidth measurements - old bonding: 8 gbps (variable due to packet reorderings) - new bonding: 10 gbps (variable due to packet reorderings) Of course the latency has improved in all converted modes, and moreover while doing enslave/release (since it doesn't affect tx anymore). Also I've stress tested all modes doing enslave/release in a loop while transmitting traffic. Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-01 18:54:51 +04:00
#ifdef CONFIG_NET_POLL_CONTROLLER
net: Convert netpoll blocking api in bonding driver to be a counter A while back I made some changes to enable netpoll in the bonding driver. Among them was a per-cpu flag that indicated we were in a path that held locks which could cause the netpoll path to block in during tx, and as such the tx path should queue the frame for later use. This appears to have given rise to a regression. If one of those paths on which we hold the per-cpu flag yields the cpu, its possible for us to come back on a different cpu, leading to us clearing a different flag than we set. This results in odd netpoll drops, and BUG backtraces appearing in the log, as we check to make sure that we only clear set bits, and only set clear bits. I had though briefly about changing the offending paths so that they wouldn't sleep, but looking at my origional work more closely, it doesn't appear that a per-cpu flag is warranted. We alrady gate the checking of this flag on IFF_IN_NETPOLL, so we don't hit this in the normal tx case anyway. And practically speaking, the normal use case for netpoll is to only have one client anyway, so we're not going to erroneously queue netpoll frames when its actually safe to do so. As such, lets just convert that per-cpu flag to an atomic counter. It fixes the rescheduling bugs, is equivalent from a performance perspective and actually eliminates some code in the process. Tested by the reporter and myself, successfully Reported-by: Liang Zheng <lzheng@redhat.com> CC: Jay Vosburgh <fubar@us.ibm.com> CC: Andy Gospodarek <andy@greyhouse.net> CC: David S. Miller <davem@davemloft.net> Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-12-06 12:05:50 +03:00
extern atomic_t netpoll_block_tx;
static inline void block_netpoll_tx(void)
{
net: Convert netpoll blocking api in bonding driver to be a counter A while back I made some changes to enable netpoll in the bonding driver. Among them was a per-cpu flag that indicated we were in a path that held locks which could cause the netpoll path to block in during tx, and as such the tx path should queue the frame for later use. This appears to have given rise to a regression. If one of those paths on which we hold the per-cpu flag yields the cpu, its possible for us to come back on a different cpu, leading to us clearing a different flag than we set. This results in odd netpoll drops, and BUG backtraces appearing in the log, as we check to make sure that we only clear set bits, and only set clear bits. I had though briefly about changing the offending paths so that they wouldn't sleep, but looking at my origional work more closely, it doesn't appear that a per-cpu flag is warranted. We alrady gate the checking of this flag on IFF_IN_NETPOLL, so we don't hit this in the normal tx case anyway. And practically speaking, the normal use case for netpoll is to only have one client anyway, so we're not going to erroneously queue netpoll frames when its actually safe to do so. As such, lets just convert that per-cpu flag to an atomic counter. It fixes the rescheduling bugs, is equivalent from a performance perspective and actually eliminates some code in the process. Tested by the reporter and myself, successfully Reported-by: Liang Zheng <lzheng@redhat.com> CC: Jay Vosburgh <fubar@us.ibm.com> CC: Andy Gospodarek <andy@greyhouse.net> CC: David S. Miller <davem@davemloft.net> Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-12-06 12:05:50 +03:00
atomic_inc(&netpoll_block_tx);
}
static inline void unblock_netpoll_tx(void)
{
net: Convert netpoll blocking api in bonding driver to be a counter A while back I made some changes to enable netpoll in the bonding driver. Among them was a per-cpu flag that indicated we were in a path that held locks which could cause the netpoll path to block in during tx, and as such the tx path should queue the frame for later use. This appears to have given rise to a regression. If one of those paths on which we hold the per-cpu flag yields the cpu, its possible for us to come back on a different cpu, leading to us clearing a different flag than we set. This results in odd netpoll drops, and BUG backtraces appearing in the log, as we check to make sure that we only clear set bits, and only set clear bits. I had though briefly about changing the offending paths so that they wouldn't sleep, but looking at my origional work more closely, it doesn't appear that a per-cpu flag is warranted. We alrady gate the checking of this flag on IFF_IN_NETPOLL, so we don't hit this in the normal tx case anyway. And practically speaking, the normal use case for netpoll is to only have one client anyway, so we're not going to erroneously queue netpoll frames when its actually safe to do so. As such, lets just convert that per-cpu flag to an atomic counter. It fixes the rescheduling bugs, is equivalent from a performance perspective and actually eliminates some code in the process. Tested by the reporter and myself, successfully Reported-by: Liang Zheng <lzheng@redhat.com> CC: Jay Vosburgh <fubar@us.ibm.com> CC: Andy Gospodarek <andy@greyhouse.net> CC: David S. Miller <davem@davemloft.net> Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-12-06 12:05:50 +03:00
atomic_dec(&netpoll_block_tx);
}
static inline int is_netpoll_tx_blocked(struct net_device *dev)
{
if (unlikely(netpoll_tx_running(dev)))
net: Convert netpoll blocking api in bonding driver to be a counter A while back I made some changes to enable netpoll in the bonding driver. Among them was a per-cpu flag that indicated we were in a path that held locks which could cause the netpoll path to block in during tx, and as such the tx path should queue the frame for later use. This appears to have given rise to a regression. If one of those paths on which we hold the per-cpu flag yields the cpu, its possible for us to come back on a different cpu, leading to us clearing a different flag than we set. This results in odd netpoll drops, and BUG backtraces appearing in the log, as we check to make sure that we only clear set bits, and only set clear bits. I had though briefly about changing the offending paths so that they wouldn't sleep, but looking at my origional work more closely, it doesn't appear that a per-cpu flag is warranted. We alrady gate the checking of this flag on IFF_IN_NETPOLL, so we don't hit this in the normal tx case anyway. And practically speaking, the normal use case for netpoll is to only have one client anyway, so we're not going to erroneously queue netpoll frames when its actually safe to do so. As such, lets just convert that per-cpu flag to an atomic counter. It fixes the rescheduling bugs, is equivalent from a performance perspective and actually eliminates some code in the process. Tested by the reporter and myself, successfully Reported-by: Liang Zheng <lzheng@redhat.com> CC: Jay Vosburgh <fubar@us.ibm.com> CC: Andy Gospodarek <andy@greyhouse.net> CC: David S. Miller <davem@davemloft.net> Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-12-06 12:05:50 +03:00
return atomic_read(&netpoll_block_tx);
return 0;
}
#else
#define block_netpoll_tx()
#define unblock_netpoll_tx()
#define is_netpoll_tx_blocked(dev) (0)
#endif
struct bond_params {
int mode;
int xmit_policy;
int miimon;
u8 num_peer_notif;
int arp_interval;
int arp_validate;
bonding: add an option to fail when any of arp_ip_target is inaccessible Currently, we fail only when all of the ips in arp_ip_target are gone. However, in some situations we might need to fail if even one host from arp_ip_target becomes unavailable. All situations, obviously, rely on the idea that we need *completely* functional network, with all interfaces/addresses working correctly. One real world example might be: vlans on top on bond (hybrid port). If bond and vlans have ips assigned and we have their peers monitored via arp_ip_target - in case of switch misconfiguration (trunk/access port), slave driver malfunction or tagged/untagged traffic dropped on the way - we will be able to switch to another slave. Though any other configuration needs that if we need to have access to all arp_ip_targets. This patch adds this possibility by adding a new parameter - arp_all_targets (both as a module parameter and as a sysfs knob). It can be set to: 0 or any (the default) - which works exactly as it's working now - the slave is up if any of the arp_ip_targets are up. 1 or all - the slave is up if all of the arp_ip_targets are up. This parameter can be changed on the fly (via sysfs), and requires the mode to be active-backup and arp_validate to be enabled (it obeys the arp_validate config on which slaves to validate). Internally it's done through: 1) Add target_last_arp_rx[BOND_MAX_ARP_TARGETS] array to slave struct. It's an array of jiffies, meaning that slave->target_last_arp_rx[i] is the last time we've received arp from bond->params.arp_targets[i] on this slave. 2) If we successfully validate an arp from bond->params.arp_targets[i] in bond_validate_arp() - update the slave->target_last_arp_rx[i] with the current jiffies value. 3) When getting slave's last_rx via slave_last_rx(), we return the oldest time when we've received an arp from any address in bond->params.arp_targets[]. If the value of arp_all_targets == 0 - we still work the same way as before. Also, update the documentation to reflect the new parameter. v3->v4: Kill the forgotten rtnl_unlock(), rephrase the documentation part to be more clear, don't fail setting arp_all_targets if arp_validate is not set - it has no effect anyway but can be easier to set up. Also, print a warning if the last arp_ip_target is removed while the arp_interval is on, but not the arp_validate. v2->v3: Use _bh spinlock, remove useless rtnl_lock() and use jiffies for new arp_ip_target last arp, instead of slave_last_rx(). On bond_enslave(), use the same initialization value for target_last_arp_rx[] as is used for the default last_arp_rx, to avoid useless interface flaps. Also, instead of failing to remove the last arp_ip_target just print a warning - otherwise it might break existing scripts. v1->v2: Correctly handle adding/removing hosts in arp_ip_target - we need to shift/initialize all slave's target_last_arp_rx. Also, don't fail module loading on arp_all_targets misconfiguration, just disable it, and some minor style fixes. Signed-off-by: Veaceslav Falico <vfalico@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-24 13:49:34 +04:00
int arp_all_targets;
int use_carrier;
int fail_over_mac;
int updelay;
int downdelay;
int lacp_fast;
unsigned int min_links;
int ad_select;
char primary[IFNAMSIZ];
int primary_reselect;
__be32 arp_targets[BOND_MAX_ARP_TARGETS];
bonding: allow user-controlled output slave selection v2: changed bonding module version, modified to apply on top of changes from previous patch in series, and updated documentation to elaborate on multiqueue awareness that now exists in bonding driver. This patch give the user the ability to control the output slave for round-robin and active-backup bonding. Similar functionality was discussed in the past, but Jay Vosburgh indicated he would rather see a feature like this added to existing modes rather than creating a completely new mode. Jay's thoughts as well as Neil's input surrounding some of the issues with the first implementation pushed us toward a design that relied on the queue_mapping rather than skb marks. Round-robin and active-backup modes were chosen as the first users of this slave selection as they seemed like the most logical choices when considering a multi-switch environment. Round-robin mode works without any modification, but active-backup does require inclusion of the first patch in this series and setting the 'all_slaves_active' flag. This will allow reception of unicast traffic on any of the backup interfaces. This was tested with IPv4-based filters as well as VLAN-based filters with good results. More information as well as a configuration example is available in the patch to Documentation/networking/bonding.txt. Signed-off-by: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-06-02 12:40:18 +04:00
int tx_queues;
int all_slaves_active;
int resend_igmp;
int lp_interval;
int packets_per_slave;
int tlb_dynamic_lb;
reciprocal_divide: update/correction of the algorithm Jakub Zawadzki noticed that some divisions by reciprocal_divide() were not correct [1][2], which he could also show with BPF code after divisions are transformed into reciprocal_value() for runtime invariance which can be passed to reciprocal_divide() later on; reverse in BPF dump ended up with a different, off-by-one K in some situations. This has been fixed by Eric Dumazet in commit aee636c4809fa5 ("bpf: do not use reciprocal divide"). This follow-up patch improves reciprocal_value() and reciprocal_divide() to work in all cases by using Granlund and Montgomery method, so that also future use is safe and without any non-obvious side-effects. Known problems with the old implementation were that division by 1 always returned 0 and some off-by-ones when the dividend and divisor where very large. This seemed to not be problematic with its current users, as far as we can tell. Eric Dumazet checked for the slab usage, we cannot surely say so in the case of flex_array. Still, in order to fix that, we propose an extension from the original implementation from commit 6a2d7a955d8d resp. [3][4], by using the algorithm proposed in "Division by Invariant Integers Using Multiplication" [5], Torbjörn Granlund and Peter L. Montgomery, that is, pseudocode for q = n/d where q, n, d is in u32 universe: 1) Initialization: int l = ceil(log_2 d) uword m' = floor((1<<32)*((1<<l)-d)/d)+1 int sh_1 = min(l,1) int sh_2 = max(l-1,0) 2) For q = n/d, all uword: uword t = (n*m')>>32 q = (t+((n-t)>>sh_1))>>sh_2 The assembler implementation from Agner Fog [6] also helped a lot while implementing. We have tested the implementation on x86_64, ppc64, i686, s390x; on x86_64/haswell we're still half the latency compared to normal divide. Joint work with Daniel Borkmann. [1] http://www.wireshark.org/~darkjames/reciprocal-buggy.c [2] http://www.wireshark.org/~darkjames/set-and-dump-filter-k-bug.c [3] https://gmplib.org/~tege/division-paper.pdf [4] http://homepage.cs.uiowa.edu/~jones/bcd/divide.html [5] http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.1.2556 [6] http://www.agner.org/optimize/asmlib.zip Reported-by: Jakub Zawadzki <darkjames-ws@darkjames.pl> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Austin S Hemmelgarn <ahferroin7@gmail.com> Cc: linux-kernel@vger.kernel.org Cc: Jesse Gross <jesse@nicira.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: Matt Mackall <mpm@selenic.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: Christoph Lameter <cl@linux-foundation.org> Cc: Andy Gospodarek <andy@greyhouse.net> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Jakub Zawadzki <darkjames-ws@darkjames.pl> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-22 05:29:41 +04:00
struct reciprocal_value reciprocal_packets_per_slave;
u16 ad_actor_sys_prio;
u16 ad_user_port_key;
/* 2 bytes of padding : see ether_addr_equal_64bits() */
u8 ad_actor_system[ETH_ALEN + 2];
};
struct bond_parm_tbl {
char *modename;
int mode;
};
struct netdev_notify_work {
struct delayed_work work;
struct net_device *dev;
struct netdev_bonding_info bonding_info;
};
struct slave {
struct net_device *dev; /* first - useful for panic debug */
struct bonding *bond; /* our master */
int delay;
/* all three in jiffies */
unsigned long last_link_up;
unsigned long last_rx;
bonding: add an option to fail when any of arp_ip_target is inaccessible Currently, we fail only when all of the ips in arp_ip_target are gone. However, in some situations we might need to fail if even one host from arp_ip_target becomes unavailable. All situations, obviously, rely on the idea that we need *completely* functional network, with all interfaces/addresses working correctly. One real world example might be: vlans on top on bond (hybrid port). If bond and vlans have ips assigned and we have their peers monitored via arp_ip_target - in case of switch misconfiguration (trunk/access port), slave driver malfunction or tagged/untagged traffic dropped on the way - we will be able to switch to another slave. Though any other configuration needs that if we need to have access to all arp_ip_targets. This patch adds this possibility by adding a new parameter - arp_all_targets (both as a module parameter and as a sysfs knob). It can be set to: 0 or any (the default) - which works exactly as it's working now - the slave is up if any of the arp_ip_targets are up. 1 or all - the slave is up if all of the arp_ip_targets are up. This parameter can be changed on the fly (via sysfs), and requires the mode to be active-backup and arp_validate to be enabled (it obeys the arp_validate config on which slaves to validate). Internally it's done through: 1) Add target_last_arp_rx[BOND_MAX_ARP_TARGETS] array to slave struct. It's an array of jiffies, meaning that slave->target_last_arp_rx[i] is the last time we've received arp from bond->params.arp_targets[i] on this slave. 2) If we successfully validate an arp from bond->params.arp_targets[i] in bond_validate_arp() - update the slave->target_last_arp_rx[i] with the current jiffies value. 3) When getting slave's last_rx via slave_last_rx(), we return the oldest time when we've received an arp from any address in bond->params.arp_targets[]. If the value of arp_all_targets == 0 - we still work the same way as before. Also, update the documentation to reflect the new parameter. v3->v4: Kill the forgotten rtnl_unlock(), rephrase the documentation part to be more clear, don't fail setting arp_all_targets if arp_validate is not set - it has no effect anyway but can be easier to set up. Also, print a warning if the last arp_ip_target is removed while the arp_interval is on, but not the arp_validate. v2->v3: Use _bh spinlock, remove useless rtnl_lock() and use jiffies for new arp_ip_target last arp, instead of slave_last_rx(). On bond_enslave(), use the same initialization value for target_last_arp_rx[] as is used for the default last_arp_rx, to avoid useless interface flaps. Also, instead of failing to remove the last arp_ip_target just print a warning - otherwise it might break existing scripts. v1->v2: Correctly handle adding/removing hosts in arp_ip_target - we need to shift/initialize all slave's target_last_arp_rx. Also, don't fail module loading on arp_all_targets misconfiguration, just disable it, and some minor style fixes. Signed-off-by: Veaceslav Falico <vfalico@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-24 13:49:34 +04:00
unsigned long target_last_arp_rx[BOND_MAX_ARP_TARGETS];
s8 link; /* one of BOND_LINK_XXXX */
s8 link_new_state; /* one of BOND_LINK_XXXX */
s8 new_link;
u8 backup:1, /* indicates backup slave. Value corresponds with
BOND_STATE_ACTIVE and BOND_STATE_BACKUP */
bonding: Fix RTNL: assertion failed at net/core/rtnetlink.c for 802.3ad mode The problem was introduced by the commit 1d3ee88ae0d (bonding: add netlink attributes to slave link dev). The bond_set_active_slave() and bond_set_backup_slave() will use rtmsg_ifinfo to send slave's states, so these two functions should be called in RTNL. In 802.3ad mode, acquiring RTNL for the __enable_port and __disable_port cases is difficult, as those calls generally already hold the state machine lock, and cannot unconditionally call rtnl_lock because either they already hold RTNL (for calls via bond_3ad_unbind_slave) or due to the potential for deadlock with bond_3ad_adapter_speed_changed, bond_3ad_adapter_duplex_changed, bond_3ad_link_change, or bond_3ad_update_lacp_rate. All four of those are called with RTNL held, and acquire the state machine lock second. The calling contexts for __enable_port and __disable_port already hold the state machine lock, and may or may not need RTNL. According to the Jay's opinion, I don't think it is a problem that the slave don't send notify message synchronously when the status changed, normally the state machine is running every 100 ms, send the notify message at the end of the state machine if the slave's state changed should be better. I fix the problem through these steps: 1). add a new function bond_set_slave_state() which could change the slave's state and call rtmsg_ifinfo() according to the input parameters called notify. 2). Add a new slave parameter which called should_notify, if the slave's state changed and don't notify yet, the parameter will be set to 1, and then if the slave's state changed again, the param will be set to 0, it indicate that the slave's state has been restored, no need to notify any one. 3). the __enable_port and __disable_port should not call rtmsg_ifinfo in the state machine lock, any change in the state of slave could set a flag in the slave, it will indicated that an rtmsg_ifinfo should be called at the end of the state machine. Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-26 07:05:22 +04:00
inactive:1, /* indicates inactive slave */
should_notify:1, /* indicates whether the state changed */
should_notify_link:1; /* indicates whether the link changed */
u8 duplex;
u32 original_mtu;
u32 link_failure_count;
u32 speed;
bonding: allow user-controlled output slave selection v2: changed bonding module version, modified to apply on top of changes from previous patch in series, and updated documentation to elaborate on multiqueue awareness that now exists in bonding driver. This patch give the user the ability to control the output slave for round-robin and active-backup bonding. Similar functionality was discussed in the past, but Jay Vosburgh indicated he would rather see a feature like this added to existing modes rather than creating a completely new mode. Jay's thoughts as well as Neil's input surrounding some of the issues with the first implementation pushed us toward a design that relied on the queue_mapping rather than skb marks. Round-robin and active-backup modes were chosen as the first users of this slave selection as they seemed like the most logical choices when considering a multi-switch environment. Round-robin mode works without any modification, but active-backup does require inclusion of the first patch in this series and setting the 'all_slaves_active' flag. This will allow reception of unicast traffic on any of the backup interfaces. This was tested with IPv4-based filters as well as VLAN-based filters with good results. More information as well as a configuration example is available in the patch to Documentation/networking/bonding.txt. Signed-off-by: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-06-02 12:40:18 +04:00
u16 queue_id;
bonding: attempt to better support longer hw addresses People are using bonding over Infiniband IPoIB connections, and who knows what else. Infiniband has a hardware address length of 20 octets (INFINIBAND_ALEN), and the network core defines a MAX_ADDR_LEN of 32. Various places in the bonding code are currently hard-wired to 6 octets (ETH_ALEN), such as the 3ad code, which I've left untouched here. Besides, only alb is currently possible on Infiniband links right now anyway, due to commit 1533e7731522, so the alb code is where most of the changes are. One major component of this change is the addition of a bond_hw_addr_copy function that takes a length argument, instead of using ether_addr_copy everywhere that hardware addresses need to be copied about. The other major component of this change is converting the bonding code from using struct sockaddr for address storage to struct sockaddr_storage, as the former has an address storage space of only 14, while the latter is 128 minus a few, which is necessary to support bonding over device with up to MAX_ADDR_LEN octet hardware addresses. Additionally, this probably fixes up some memory corruption issues with the current code, where it's possible to write an infiniband hardware address into a sockaddr declared on the stack. Lightly tested on a dual mlx4 IPoIB setup, which properly shows a 20-octet hardware address now: $ cat /proc/net/bonding/bond0 Ethernet Channel Bonding Driver: v3.7.1 (April 27, 2011) Bonding Mode: fault-tolerance (active-backup) (fail_over_mac active) Primary Slave: mlx4_ib0 (primary_reselect always) Currently Active Slave: mlx4_ib0 MII Status: up MII Polling Interval (ms): 100 Up Delay (ms): 100 Down Delay (ms): 100 Slave Interface: mlx4_ib0 MII Status: up Speed: Unknown Duplex: Unknown Link Failure Count: 0 Permanent HW addr: 80:00:02:08:fe:80:00:00:00:00:00:00:e4:1d:2d:03:00:1d:67:01 Slave queue ID: 0 Slave Interface: mlx4_ib1 MII Status: up Speed: Unknown Duplex: Unknown Link Failure Count: 0 Permanent HW addr: 80:00:02:09:fe:80:00:00:00:00:00:01:e4:1d:2d:03:00:1d:67:02 Slave queue ID: 0 Also tested with a standard 1Gbps NIC bonding setup (with a mix of e1000 and e1000e cards), running LNST's bonding tests. CC: Jay Vosburgh <j.vosburgh@gmail.com> CC: Veaceslav Falico <vfalico@gmail.com> CC: Andy Gospodarek <andy@greyhouse.net> CC: netdev@vger.kernel.org Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-05 00:32:42 +03:00
u8 perm_hwaddr[MAX_ADDR_LEN];
struct ad_slave_info *ad_info;
struct tlb_slave_info tlb_info;
#ifdef CONFIG_NET_POLL_CONTROLLER
struct netpoll *np;
#endif
struct kobject kobj;
struct rtnl_link_stats64 slave_stats;
};
struct bond_up_slave {
unsigned int count;
struct rcu_head rcu;
struct slave *arr[0];
};
/*
* Link pseudo-state only used internally by monitors
*/
#define BOND_LINK_NOCHANGE -1
/*
* Here are the locking policies for the two bonding locks:
* Get rcu_read_lock when reading or RTNL when writing slave list.
*/
struct bonding {
struct net_device *dev; /* first - useful for panic debug */
struct slave __rcu *curr_active_slave;
struct slave __rcu *current_arp_slave;
struct slave __rcu *primary_slave;
struct bond_up_slave __rcu *slave_arr; /* Array of usable slaves */
bool force_primary;
s32 slave_cnt; /* never change this value outside the attach/detach wrappers */
int (*recv_probe)(const struct sk_buff *, struct bonding *,
struct slave *);
/* mode_lock is used for mode-specific locking needs, currently used by:
* 3ad mode (4) - protect against running bond_3ad_unbind_slave() and
* bond_3ad_state_machine_handler() concurrently and also
* the access to the state machine shared variables.
* TLB mode (5) - to sync the use and modifications of its hash table
* ALB mode (6) - to sync the use and modifications of its hash table
*/
spinlock_t mode_lock;
spinlock_t stats_lock;
u8 send_peer_notif;
bonding: fix igmp_retrans type and two related races First the type of igmp_retrans (which is the actual counter of igmp_resend parameter) is changed to u8 to be able to store values up to 255 (as per documentation). There are two races that were hidden there and which are easy to trigger after the previous fix, the first is between bond_resend_igmp_join_requests and bond_change_active_slave where igmp_retrans is set and can be altered by the periodic. The second race condition is between multiple running instances of the periodic (upon execution it can be scheduled again for immediate execution which can cause the counter to go < 0 which in the unsigned case leads to unnecessary igmp retransmissions). Since in bond_change_active_slave bond->lock is held for reading and curr_slave_lock for writing, we use curr_slave_lock for mutual exclusion. We can't drop them as there're cases where RTNL is not held when bond_change_active_slave is called. RCU is unlocked in bond_resend_igmp_join_requests before getting curr_slave_lock since we don't need it there and it's pointless to delay. The decrement is moved inside the "if" block because if we decrement unconditionally there's still a possibility for a race condition although it is much more difficult to hit (many changes have to happen in a very short period in order to trigger) which in the case of 3 parallel running instances of this function and igmp_retrans == 1 (with check bond->igmp_retrans-- > 1) is: f1 passes, doesn't re-schedule, but decrements - igmp_retrans = 0 f2 then passes, doesn't re-schedule, but decrements - igmp_retrans = 255 f3 does the unnecessary retransmissions. Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com> Signed-off-by: Jay Vosburgh <fubar@us.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-12 02:07:02 +04:00
u8 igmp_retrans;
#ifdef CONFIG_PROC_FS
struct proc_dir_entry *proc_entry;
char proc_file_name[IFNAMSIZ];
#endif /* CONFIG_PROC_FS */
struct list_head bond_list;
u32 rr_tx_counter;
struct ad_bond_info ad_info;
struct alb_bond_info alb_info;
struct bond_params params;
struct workqueue_struct *wq;
struct delayed_work mii_work;
struct delayed_work arp_work;
struct delayed_work alb_work;
struct delayed_work ad_work;
struct delayed_work mcast_work;
struct delayed_work slave_arr_work;
#ifdef CONFIG_DEBUG_FS
/* debugging support via debugfs */
struct dentry *debug_dir;
#endif /* CONFIG_DEBUG_FS */
struct rtnl_link_stats64 bond_stats;
};
#define bond_slave_get_rcu(dev) \
((struct slave *) rcu_dereference(dev->rx_handler_data))
#define bond_slave_get_rtnl(dev) \
((struct slave *) rtnl_dereference(dev->rx_handler_data))
void bond_queue_slave_event(struct slave *slave);
void bond_lower_state_changed(struct slave *slave);
bonding: support QinQ for bond arp interval The bond send arp request to indicate that the slave is active, and if the bond dev is a vlan dev, it will set the vlan tag in skb to notice the vlan group, but the bond could only send a skb with 802.1q proto, not support for QinQ. So add outer tag for lower vlan tag and inner tag for upper vlan tag to support QinQ, The new skb will be consist of two vlan tag just like this: dst mac | src mac | outer vlan tag | inner vlan tag | data | ..... If We don't need QinQ, the inner vlan tag could be set to 0 and use outer vlan tag as a normal vlan group. Using "ip link" to configure the bond for QinQ and add test log: ip link add link bond0 bond0.20 type vlan proto 802.1ad id 20 ip link add link bond0.20 bond0.20.200 type vlan proto 802.1q id 200 ifconfig bond0.20 11.11.20.36/24 ifconfig bond0.20.200 11.11.200.36/24 echo +11.11.200.37 > /sys/class/net/bond0/bonding/arp_ip_target 90:e2:ba:07:4a:5c (oui Unknown) > Broadcast, ethertype 802.1Q-QinQ (0x88a8),length 50: vlan 20, p 0,ethertype 802.1Q, vlan 200, p 0, ethertype ARP, Ethernet (len 6), IPv4 (len 4), Request who-has 11.11.200.37 tell 11.11.200.36, length 28 90:e2:ba:06:f9:86 (oui Unknown) > 90:e2:ba:07:4a:5c (oui Unknown), ethertype 802.1Q-QinQ (0x88a8), length 50: vlan 20, p 0, ethertype 802.1Q, vlan 200, p 0, ethertype ARP, Ethernet (len 6), IPv4 (len 4), Reply 11.11.200.37 is-at 90:e2:ba:06:f9:86 (oui Unknown), length 28 v1->v2: remove the comment "TODO: QinQ?". Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-25 13:44:43 +04:00
struct bond_vlan_tag {
__be16 vlan_proto;
unsigned short vlan_id;
};
/**
* Returns NULL if the net_device does not belong to any of the bond's slaves
*
* Caller must hold bond lock for read
*/
static inline struct slave *bond_get_slave_by_dev(struct bonding *bond,
struct net_device *slave_dev)
{
return netdev_lower_dev_get_private(bond->dev, slave_dev);
}
static inline struct bonding *bond_get_bond_by_slave(struct slave *slave)
{
return slave->bond;
}
static inline bool bond_should_override_tx_queue(struct bonding *bond)
{
return BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP ||
BOND_MODE(bond) == BOND_MODE_ROUNDROBIN;
}
static inline bool bond_is_lb(const struct bonding *bond)
{
return BOND_MODE(bond) == BOND_MODE_TLB ||
BOND_MODE(bond) == BOND_MODE_ALB;
}
static inline bool bond_is_nondyn_tlb(const struct bonding *bond)
{
return (BOND_MODE(bond) == BOND_MODE_TLB) &&
(bond->params.tlb_dynamic_lb == 0);
}
static inline bool bond_mode_uses_xmit_hash(const struct bonding *bond)
{
return (BOND_MODE(bond) == BOND_MODE_8023AD ||
BOND_MODE(bond) == BOND_MODE_XOR ||
bond_is_nondyn_tlb(bond));
}
static inline bool bond_mode_uses_arp(int mode)
{
return mode != BOND_MODE_8023AD && mode != BOND_MODE_TLB &&
mode != BOND_MODE_ALB;
}
static inline bool bond_mode_uses_primary(int mode)
{
return mode == BOND_MODE_ACTIVEBACKUP || mode == BOND_MODE_TLB ||
mode == BOND_MODE_ALB;
}
static inline bool bond_uses_primary(struct bonding *bond)
{
return bond_mode_uses_primary(BOND_MODE(bond));
}
static inline struct net_device *bond_option_active_slave_get_rcu(struct bonding *bond)
{
struct slave *slave = rcu_dereference(bond->curr_active_slave);
return bond_uses_primary(bond) && slave ? slave->dev : NULL;
}
static inline bool bond_slave_is_up(struct slave *slave)
{
return netif_running(slave->dev) && netif_carrier_ok(slave->dev);
}
static inline void bond_set_active_slave(struct slave *slave)
{
if (slave->backup) {
slave->backup = 0;
bond_queue_slave_event(slave);
bond_lower_state_changed(slave);
rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC);
}
}
static inline void bond_set_backup_slave(struct slave *slave)
{
if (!slave->backup) {
slave->backup = 1;
bond_queue_slave_event(slave);
bond_lower_state_changed(slave);
rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC);
}
}
bonding: Fix RTNL: assertion failed at net/core/rtnetlink.c for 802.3ad mode The problem was introduced by the commit 1d3ee88ae0d (bonding: add netlink attributes to slave link dev). The bond_set_active_slave() and bond_set_backup_slave() will use rtmsg_ifinfo to send slave's states, so these two functions should be called in RTNL. In 802.3ad mode, acquiring RTNL for the __enable_port and __disable_port cases is difficult, as those calls generally already hold the state machine lock, and cannot unconditionally call rtnl_lock because either they already hold RTNL (for calls via bond_3ad_unbind_slave) or due to the potential for deadlock with bond_3ad_adapter_speed_changed, bond_3ad_adapter_duplex_changed, bond_3ad_link_change, or bond_3ad_update_lacp_rate. All four of those are called with RTNL held, and acquire the state machine lock second. The calling contexts for __enable_port and __disable_port already hold the state machine lock, and may or may not need RTNL. According to the Jay's opinion, I don't think it is a problem that the slave don't send notify message synchronously when the status changed, normally the state machine is running every 100 ms, send the notify message at the end of the state machine if the slave's state changed should be better. I fix the problem through these steps: 1). add a new function bond_set_slave_state() which could change the slave's state and call rtmsg_ifinfo() according to the input parameters called notify. 2). Add a new slave parameter which called should_notify, if the slave's state changed and don't notify yet, the parameter will be set to 1, and then if the slave's state changed again, the param will be set to 0, it indicate that the slave's state has been restored, no need to notify any one. 3). the __enable_port and __disable_port should not call rtmsg_ifinfo in the state machine lock, any change in the state of slave could set a flag in the slave, it will indicated that an rtmsg_ifinfo should be called at the end of the state machine. Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-26 07:05:22 +04:00
static inline void bond_set_slave_state(struct slave *slave,
int slave_state, bool notify)
{
if (slave->backup == slave_state)
return;
slave->backup = slave_state;
if (notify) {
bond_lower_state_changed(slave);
rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC);
bond_queue_slave_event(slave);
bonding: Fix RTNL: assertion failed at net/core/rtnetlink.c for 802.3ad mode The problem was introduced by the commit 1d3ee88ae0d (bonding: add netlink attributes to slave link dev). The bond_set_active_slave() and bond_set_backup_slave() will use rtmsg_ifinfo to send slave's states, so these two functions should be called in RTNL. In 802.3ad mode, acquiring RTNL for the __enable_port and __disable_port cases is difficult, as those calls generally already hold the state machine lock, and cannot unconditionally call rtnl_lock because either they already hold RTNL (for calls via bond_3ad_unbind_slave) or due to the potential for deadlock with bond_3ad_adapter_speed_changed, bond_3ad_adapter_duplex_changed, bond_3ad_link_change, or bond_3ad_update_lacp_rate. All four of those are called with RTNL held, and acquire the state machine lock second. The calling contexts for __enable_port and __disable_port already hold the state machine lock, and may or may not need RTNL. According to the Jay's opinion, I don't think it is a problem that the slave don't send notify message synchronously when the status changed, normally the state machine is running every 100 ms, send the notify message at the end of the state machine if the slave's state changed should be better. I fix the problem through these steps: 1). add a new function bond_set_slave_state() which could change the slave's state and call rtmsg_ifinfo() according to the input parameters called notify. 2). Add a new slave parameter which called should_notify, if the slave's state changed and don't notify yet, the parameter will be set to 1, and then if the slave's state changed again, the param will be set to 0, it indicate that the slave's state has been restored, no need to notify any one. 3). the __enable_port and __disable_port should not call rtmsg_ifinfo in the state machine lock, any change in the state of slave could set a flag in the slave, it will indicated that an rtmsg_ifinfo should be called at the end of the state machine. Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-26 07:05:22 +04:00
slave->should_notify = 0;
} else {
if (slave->should_notify)
slave->should_notify = 0;
else
slave->should_notify = 1;
}
}
static inline void bond_slave_state_change(struct bonding *bond)
{
struct list_head *iter;
struct slave *tmp;
bond_for_each_slave(bond, tmp, iter) {
if (tmp->link == BOND_LINK_UP)
bond_set_active_slave(tmp);
else if (tmp->link == BOND_LINK_DOWN)
bond_set_backup_slave(tmp);
}
}
static inline void bond_slave_state_notify(struct bonding *bond)
{
struct list_head *iter;
struct slave *tmp;
bond_for_each_slave(bond, tmp, iter) {
if (tmp->should_notify) {
bond_lower_state_changed(tmp);
rtmsg_ifinfo(RTM_NEWLINK, tmp->dev, 0, GFP_ATOMIC);
tmp->should_notify = 0;
}
}
}
static inline int bond_slave_state(struct slave *slave)
{
return slave->backup;
}
static inline bool bond_is_active_slave(struct slave *slave)
{
return !bond_slave_state(slave);
}
static inline bool bond_slave_can_tx(struct slave *slave)
{
return bond_slave_is_up(slave) && slave->link == BOND_LINK_UP &&
bond_is_active_slave(slave);
}
bonding: attempt to better support longer hw addresses People are using bonding over Infiniband IPoIB connections, and who knows what else. Infiniband has a hardware address length of 20 octets (INFINIBAND_ALEN), and the network core defines a MAX_ADDR_LEN of 32. Various places in the bonding code are currently hard-wired to 6 octets (ETH_ALEN), such as the 3ad code, which I've left untouched here. Besides, only alb is currently possible on Infiniband links right now anyway, due to commit 1533e7731522, so the alb code is where most of the changes are. One major component of this change is the addition of a bond_hw_addr_copy function that takes a length argument, instead of using ether_addr_copy everywhere that hardware addresses need to be copied about. The other major component of this change is converting the bonding code from using struct sockaddr for address storage to struct sockaddr_storage, as the former has an address storage space of only 14, while the latter is 128 minus a few, which is necessary to support bonding over device with up to MAX_ADDR_LEN octet hardware addresses. Additionally, this probably fixes up some memory corruption issues with the current code, where it's possible to write an infiniband hardware address into a sockaddr declared on the stack. Lightly tested on a dual mlx4 IPoIB setup, which properly shows a 20-octet hardware address now: $ cat /proc/net/bonding/bond0 Ethernet Channel Bonding Driver: v3.7.1 (April 27, 2011) Bonding Mode: fault-tolerance (active-backup) (fail_over_mac active) Primary Slave: mlx4_ib0 (primary_reselect always) Currently Active Slave: mlx4_ib0 MII Status: up MII Polling Interval (ms): 100 Up Delay (ms): 100 Down Delay (ms): 100 Slave Interface: mlx4_ib0 MII Status: up Speed: Unknown Duplex: Unknown Link Failure Count: 0 Permanent HW addr: 80:00:02:08:fe:80:00:00:00:00:00:00:e4:1d:2d:03:00:1d:67:01 Slave queue ID: 0 Slave Interface: mlx4_ib1 MII Status: up Speed: Unknown Duplex: Unknown Link Failure Count: 0 Permanent HW addr: 80:00:02:09:fe:80:00:00:00:00:00:01:e4:1d:2d:03:00:1d:67:02 Slave queue ID: 0 Also tested with a standard 1Gbps NIC bonding setup (with a mix of e1000 and e1000e cards), running LNST's bonding tests. CC: Jay Vosburgh <j.vosburgh@gmail.com> CC: Veaceslav Falico <vfalico@gmail.com> CC: Andy Gospodarek <andy@greyhouse.net> CC: netdev@vger.kernel.org Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-05 00:32:42 +03:00
static inline void bond_hw_addr_copy(u8 *dst, const u8 *src, unsigned int len)
{
if (len == ETH_ALEN) {
ether_addr_copy(dst, src);
return;
}
memcpy(dst, src, len);
}
#define BOND_PRI_RESELECT_ALWAYS 0
#define BOND_PRI_RESELECT_BETTER 1
#define BOND_PRI_RESELECT_FAILURE 2
#define BOND_FOM_NONE 0
#define BOND_FOM_ACTIVE 1
#define BOND_FOM_FOLLOW 2
bonding: add an option to fail when any of arp_ip_target is inaccessible Currently, we fail only when all of the ips in arp_ip_target are gone. However, in some situations we might need to fail if even one host from arp_ip_target becomes unavailable. All situations, obviously, rely on the idea that we need *completely* functional network, with all interfaces/addresses working correctly. One real world example might be: vlans on top on bond (hybrid port). If bond and vlans have ips assigned and we have their peers monitored via arp_ip_target - in case of switch misconfiguration (trunk/access port), slave driver malfunction or tagged/untagged traffic dropped on the way - we will be able to switch to another slave. Though any other configuration needs that if we need to have access to all arp_ip_targets. This patch adds this possibility by adding a new parameter - arp_all_targets (both as a module parameter and as a sysfs knob). It can be set to: 0 or any (the default) - which works exactly as it's working now - the slave is up if any of the arp_ip_targets are up. 1 or all - the slave is up if all of the arp_ip_targets are up. This parameter can be changed on the fly (via sysfs), and requires the mode to be active-backup and arp_validate to be enabled (it obeys the arp_validate config on which slaves to validate). Internally it's done through: 1) Add target_last_arp_rx[BOND_MAX_ARP_TARGETS] array to slave struct. It's an array of jiffies, meaning that slave->target_last_arp_rx[i] is the last time we've received arp from bond->params.arp_targets[i] on this slave. 2) If we successfully validate an arp from bond->params.arp_targets[i] in bond_validate_arp() - update the slave->target_last_arp_rx[i] with the current jiffies value. 3) When getting slave's last_rx via slave_last_rx(), we return the oldest time when we've received an arp from any address in bond->params.arp_targets[]. If the value of arp_all_targets == 0 - we still work the same way as before. Also, update the documentation to reflect the new parameter. v3->v4: Kill the forgotten rtnl_unlock(), rephrase the documentation part to be more clear, don't fail setting arp_all_targets if arp_validate is not set - it has no effect anyway but can be easier to set up. Also, print a warning if the last arp_ip_target is removed while the arp_interval is on, but not the arp_validate. v2->v3: Use _bh spinlock, remove useless rtnl_lock() and use jiffies for new arp_ip_target last arp, instead of slave_last_rx(). On bond_enslave(), use the same initialization value for target_last_arp_rx[] as is used for the default last_arp_rx, to avoid useless interface flaps. Also, instead of failing to remove the last arp_ip_target just print a warning - otherwise it might break existing scripts. v1->v2: Correctly handle adding/removing hosts in arp_ip_target - we need to shift/initialize all slave's target_last_arp_rx. Also, don't fail module loading on arp_all_targets misconfiguration, just disable it, and some minor style fixes. Signed-off-by: Veaceslav Falico <vfalico@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-24 13:49:34 +04:00
#define BOND_ARP_TARGETS_ANY 0
#define BOND_ARP_TARGETS_ALL 1
#define BOND_ARP_VALIDATE_NONE 0
#define BOND_ARP_VALIDATE_ACTIVE (1 << BOND_STATE_ACTIVE)
#define BOND_ARP_VALIDATE_BACKUP (1 << BOND_STATE_BACKUP)
#define BOND_ARP_VALIDATE_ALL (BOND_ARP_VALIDATE_ACTIVE | \
BOND_ARP_VALIDATE_BACKUP)
#define BOND_ARP_FILTER (BOND_ARP_VALIDATE_ALL + 1)
#define BOND_ARP_FILTER_ACTIVE (BOND_ARP_VALIDATE_ACTIVE | \
BOND_ARP_FILTER)
#define BOND_ARP_FILTER_BACKUP (BOND_ARP_VALIDATE_BACKUP | \
BOND_ARP_FILTER)
bonding: Fix RTNL: assertion failed at net/core/rtnetlink.c for 802.3ad mode The problem was introduced by the commit 1d3ee88ae0d (bonding: add netlink attributes to slave link dev). The bond_set_active_slave() and bond_set_backup_slave() will use rtmsg_ifinfo to send slave's states, so these two functions should be called in RTNL. In 802.3ad mode, acquiring RTNL for the __enable_port and __disable_port cases is difficult, as those calls generally already hold the state machine lock, and cannot unconditionally call rtnl_lock because either they already hold RTNL (for calls via bond_3ad_unbind_slave) or due to the potential for deadlock with bond_3ad_adapter_speed_changed, bond_3ad_adapter_duplex_changed, bond_3ad_link_change, or bond_3ad_update_lacp_rate. All four of those are called with RTNL held, and acquire the state machine lock second. The calling contexts for __enable_port and __disable_port already hold the state machine lock, and may or may not need RTNL. According to the Jay's opinion, I don't think it is a problem that the slave don't send notify message synchronously when the status changed, normally the state machine is running every 100 ms, send the notify message at the end of the state machine if the slave's state changed should be better. I fix the problem through these steps: 1). add a new function bond_set_slave_state() which could change the slave's state and call rtmsg_ifinfo() according to the input parameters called notify. 2). Add a new slave parameter which called should_notify, if the slave's state changed and don't notify yet, the parameter will be set to 1, and then if the slave's state changed again, the param will be set to 0, it indicate that the slave's state has been restored, no need to notify any one. 3). the __enable_port and __disable_port should not call rtmsg_ifinfo in the state machine lock, any change in the state of slave could set a flag in the slave, it will indicated that an rtmsg_ifinfo should be called at the end of the state machine. Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-26 07:05:22 +04:00
#define BOND_SLAVE_NOTIFY_NOW true
#define BOND_SLAVE_NOTIFY_LATER false
static inline int slave_do_arp_validate(struct bonding *bond,
struct slave *slave)
{
return bond->params.arp_validate & (1 << bond_slave_state(slave));
}
static inline int slave_do_arp_validate_only(struct bonding *bond)
{
return bond->params.arp_validate & BOND_ARP_FILTER;
}
static inline int bond_is_ip_target_ok(__be32 addr)
{
return !ipv4_is_lbcast(addr) && !ipv4_is_zeronet(addr);
}
bonding: add an option to fail when any of arp_ip_target is inaccessible Currently, we fail only when all of the ips in arp_ip_target are gone. However, in some situations we might need to fail if even one host from arp_ip_target becomes unavailable. All situations, obviously, rely on the idea that we need *completely* functional network, with all interfaces/addresses working correctly. One real world example might be: vlans on top on bond (hybrid port). If bond and vlans have ips assigned and we have their peers monitored via arp_ip_target - in case of switch misconfiguration (trunk/access port), slave driver malfunction or tagged/untagged traffic dropped on the way - we will be able to switch to another slave. Though any other configuration needs that if we need to have access to all arp_ip_targets. This patch adds this possibility by adding a new parameter - arp_all_targets (both as a module parameter and as a sysfs knob). It can be set to: 0 or any (the default) - which works exactly as it's working now - the slave is up if any of the arp_ip_targets are up. 1 or all - the slave is up if all of the arp_ip_targets are up. This parameter can be changed on the fly (via sysfs), and requires the mode to be active-backup and arp_validate to be enabled (it obeys the arp_validate config on which slaves to validate). Internally it's done through: 1) Add target_last_arp_rx[BOND_MAX_ARP_TARGETS] array to slave struct. It's an array of jiffies, meaning that slave->target_last_arp_rx[i] is the last time we've received arp from bond->params.arp_targets[i] on this slave. 2) If we successfully validate an arp from bond->params.arp_targets[i] in bond_validate_arp() - update the slave->target_last_arp_rx[i] with the current jiffies value. 3) When getting slave's last_rx via slave_last_rx(), we return the oldest time when we've received an arp from any address in bond->params.arp_targets[]. If the value of arp_all_targets == 0 - we still work the same way as before. Also, update the documentation to reflect the new parameter. v3->v4: Kill the forgotten rtnl_unlock(), rephrase the documentation part to be more clear, don't fail setting arp_all_targets if arp_validate is not set - it has no effect anyway but can be easier to set up. Also, print a warning if the last arp_ip_target is removed while the arp_interval is on, but not the arp_validate. v2->v3: Use _bh spinlock, remove useless rtnl_lock() and use jiffies for new arp_ip_target last arp, instead of slave_last_rx(). On bond_enslave(), use the same initialization value for target_last_arp_rx[] as is used for the default last_arp_rx, to avoid useless interface flaps. Also, instead of failing to remove the last arp_ip_target just print a warning - otherwise it might break existing scripts. v1->v2: Correctly handle adding/removing hosts in arp_ip_target - we need to shift/initialize all slave's target_last_arp_rx. Also, don't fail module loading on arp_all_targets misconfiguration, just disable it, and some minor style fixes. Signed-off-by: Veaceslav Falico <vfalico@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-24 13:49:34 +04:00
/* Get the oldest arp which we've received on this slave for bond's
* arp_targets.
*/
static inline unsigned long slave_oldest_target_arp_rx(struct bonding *bond,
struct slave *slave)
{
int i = 1;
unsigned long ret = slave->target_last_arp_rx[0];
for (; (i < BOND_MAX_ARP_TARGETS) && bond->params.arp_targets[i]; i++)
if (time_before(slave->target_last_arp_rx[i], ret))
ret = slave->target_last_arp_rx[i];
return ret;
}
static inline unsigned long slave_last_rx(struct bonding *bond,
struct slave *slave)
{
if (bond->params.arp_all_targets == BOND_ARP_TARGETS_ALL)
return slave_oldest_target_arp_rx(bond, slave);
return slave->last_rx;
}
#ifdef CONFIG_NET_POLL_CONTROLLER
static inline void bond_netpoll_send_skb(const struct slave *slave,
struct sk_buff *skb)
{
struct netpoll *np = slave->np;
if (np)
netpoll_send_skb(np, skb);
}
#else
static inline void bond_netpoll_send_skb(const struct slave *slave,
struct sk_buff *skb)
{
}
#endif
bonding: Fix RTNL: assertion failed at net/core/rtnetlink.c for 802.3ad mode The problem was introduced by the commit 1d3ee88ae0d (bonding: add netlink attributes to slave link dev). The bond_set_active_slave() and bond_set_backup_slave() will use rtmsg_ifinfo to send slave's states, so these two functions should be called in RTNL. In 802.3ad mode, acquiring RTNL for the __enable_port and __disable_port cases is difficult, as those calls generally already hold the state machine lock, and cannot unconditionally call rtnl_lock because either they already hold RTNL (for calls via bond_3ad_unbind_slave) or due to the potential for deadlock with bond_3ad_adapter_speed_changed, bond_3ad_adapter_duplex_changed, bond_3ad_link_change, or bond_3ad_update_lacp_rate. All four of those are called with RTNL held, and acquire the state machine lock second. The calling contexts for __enable_port and __disable_port already hold the state machine lock, and may or may not need RTNL. According to the Jay's opinion, I don't think it is a problem that the slave don't send notify message synchronously when the status changed, normally the state machine is running every 100 ms, send the notify message at the end of the state machine if the slave's state changed should be better. I fix the problem through these steps: 1). add a new function bond_set_slave_state() which could change the slave's state and call rtmsg_ifinfo() according to the input parameters called notify. 2). Add a new slave parameter which called should_notify, if the slave's state changed and don't notify yet, the parameter will be set to 1, and then if the slave's state changed again, the param will be set to 0, it indicate that the slave's state has been restored, no need to notify any one. 3). the __enable_port and __disable_port should not call rtmsg_ifinfo in the state machine lock, any change in the state of slave could set a flag in the slave, it will indicated that an rtmsg_ifinfo should be called at the end of the state machine. Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-26 07:05:22 +04:00
static inline void bond_set_slave_inactive_flags(struct slave *slave,
bool notify)
{
if (!bond_is_lb(slave->bond))
bonding: Fix RTNL: assertion failed at net/core/rtnetlink.c for 802.3ad mode The problem was introduced by the commit 1d3ee88ae0d (bonding: add netlink attributes to slave link dev). The bond_set_active_slave() and bond_set_backup_slave() will use rtmsg_ifinfo to send slave's states, so these two functions should be called in RTNL. In 802.3ad mode, acquiring RTNL for the __enable_port and __disable_port cases is difficult, as those calls generally already hold the state machine lock, and cannot unconditionally call rtnl_lock because either they already hold RTNL (for calls via bond_3ad_unbind_slave) or due to the potential for deadlock with bond_3ad_adapter_speed_changed, bond_3ad_adapter_duplex_changed, bond_3ad_link_change, or bond_3ad_update_lacp_rate. All four of those are called with RTNL held, and acquire the state machine lock second. The calling contexts for __enable_port and __disable_port already hold the state machine lock, and may or may not need RTNL. According to the Jay's opinion, I don't think it is a problem that the slave don't send notify message synchronously when the status changed, normally the state machine is running every 100 ms, send the notify message at the end of the state machine if the slave's state changed should be better. I fix the problem through these steps: 1). add a new function bond_set_slave_state() which could change the slave's state and call rtmsg_ifinfo() according to the input parameters called notify. 2). Add a new slave parameter which called should_notify, if the slave's state changed and don't notify yet, the parameter will be set to 1, and then if the slave's state changed again, the param will be set to 0, it indicate that the slave's state has been restored, no need to notify any one. 3). the __enable_port and __disable_port should not call rtmsg_ifinfo in the state machine lock, any change in the state of slave could set a flag in the slave, it will indicated that an rtmsg_ifinfo should be called at the end of the state machine. Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-26 07:05:22 +04:00
bond_set_slave_state(slave, BOND_STATE_BACKUP, notify);
if (!slave->bond->params.all_slaves_active)
slave->inactive = 1;
}
bonding: Fix RTNL: assertion failed at net/core/rtnetlink.c for 802.3ad mode The problem was introduced by the commit 1d3ee88ae0d (bonding: add netlink attributes to slave link dev). The bond_set_active_slave() and bond_set_backup_slave() will use rtmsg_ifinfo to send slave's states, so these two functions should be called in RTNL. In 802.3ad mode, acquiring RTNL for the __enable_port and __disable_port cases is difficult, as those calls generally already hold the state machine lock, and cannot unconditionally call rtnl_lock because either they already hold RTNL (for calls via bond_3ad_unbind_slave) or due to the potential for deadlock with bond_3ad_adapter_speed_changed, bond_3ad_adapter_duplex_changed, bond_3ad_link_change, or bond_3ad_update_lacp_rate. All four of those are called with RTNL held, and acquire the state machine lock second. The calling contexts for __enable_port and __disable_port already hold the state machine lock, and may or may not need RTNL. According to the Jay's opinion, I don't think it is a problem that the slave don't send notify message synchronously when the status changed, normally the state machine is running every 100 ms, send the notify message at the end of the state machine if the slave's state changed should be better. I fix the problem through these steps: 1). add a new function bond_set_slave_state() which could change the slave's state and call rtmsg_ifinfo() according to the input parameters called notify. 2). Add a new slave parameter which called should_notify, if the slave's state changed and don't notify yet, the parameter will be set to 1, and then if the slave's state changed again, the param will be set to 0, it indicate that the slave's state has been restored, no need to notify any one. 3). the __enable_port and __disable_port should not call rtmsg_ifinfo in the state machine lock, any change in the state of slave could set a flag in the slave, it will indicated that an rtmsg_ifinfo should be called at the end of the state machine. Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-26 07:05:22 +04:00
static inline void bond_set_slave_active_flags(struct slave *slave,
bool notify)
{
bonding: Fix RTNL: assertion failed at net/core/rtnetlink.c for 802.3ad mode The problem was introduced by the commit 1d3ee88ae0d (bonding: add netlink attributes to slave link dev). The bond_set_active_slave() and bond_set_backup_slave() will use rtmsg_ifinfo to send slave's states, so these two functions should be called in RTNL. In 802.3ad mode, acquiring RTNL for the __enable_port and __disable_port cases is difficult, as those calls generally already hold the state machine lock, and cannot unconditionally call rtnl_lock because either they already hold RTNL (for calls via bond_3ad_unbind_slave) or due to the potential for deadlock with bond_3ad_adapter_speed_changed, bond_3ad_adapter_duplex_changed, bond_3ad_link_change, or bond_3ad_update_lacp_rate. All four of those are called with RTNL held, and acquire the state machine lock second. The calling contexts for __enable_port and __disable_port already hold the state machine lock, and may or may not need RTNL. According to the Jay's opinion, I don't think it is a problem that the slave don't send notify message synchronously when the status changed, normally the state machine is running every 100 ms, send the notify message at the end of the state machine if the slave's state changed should be better. I fix the problem through these steps: 1). add a new function bond_set_slave_state() which could change the slave's state and call rtmsg_ifinfo() according to the input parameters called notify. 2). Add a new slave parameter which called should_notify, if the slave's state changed and don't notify yet, the parameter will be set to 1, and then if the slave's state changed again, the param will be set to 0, it indicate that the slave's state has been restored, no need to notify any one. 3). the __enable_port and __disable_port should not call rtmsg_ifinfo in the state machine lock, any change in the state of slave could set a flag in the slave, it will indicated that an rtmsg_ifinfo should be called at the end of the state machine. Cc: Jay Vosburgh <fubar@us.ibm.com> Cc: Veaceslav Falico <vfalico@redhat.com> Cc: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-26 07:05:22 +04:00
bond_set_slave_state(slave, BOND_STATE_ACTIVE, notify);
slave->inactive = 0;
}
static inline bool bond_is_slave_inactive(struct slave *slave)
{
return slave->inactive;
}
static inline void bond_propose_link_state(struct slave *slave, int state)
{
slave->link_new_state = state;
}
static inline void bond_commit_link_state(struct slave *slave, bool notify)
{
if (slave->link == slave->link_new_state)
return;
slave->link = slave->link_new_state;
if (notify) {
bond_queue_slave_event(slave);
bond_lower_state_changed(slave);
slave->should_notify_link = 0;
} else {
if (slave->should_notify_link)
slave->should_notify_link = 0;
else
slave->should_notify_link = 1;
}
}
static inline void bond_set_slave_link_state(struct slave *slave, int state,
bool notify)
{
bond_propose_link_state(slave, state);
bond_commit_link_state(slave, notify);
}
static inline void bond_slave_link_notify(struct bonding *bond)
{
struct list_head *iter;
struct slave *tmp;
bond_for_each_slave(bond, tmp, iter) {
if (tmp->should_notify_link) {
bond_queue_slave_event(tmp);
bond_lower_state_changed(tmp);
tmp->should_notify_link = 0;
}
}
}
bonding: remove entries for master_ip and vlan_ip and query devices instead The following patch aimed to resolve an issue where secondary, tertiary, etc. addresses added to bond interfaces could overwrite the bond->master_ip and vlan_ip values. commit 917fbdb32f37e9a93b00bb12ee83532982982df3 Author: Henrik Saavedra Persson <henrik.e.persson@ericsson.com> Date: Wed Nov 23 23:37:15 2011 +0000 bonding: only use primary address for ARP That patch was good because it prevented bonds using ARP monitoring from sending frames with an invalid source IP address. Unfortunately, it didn't always work as expected. When using an ioctl (like ifconfig does) to set the IP address and netmask, 2 separate ioctls are actually called to set the IP and netmask if the mask chosen doesn't match the standard mask for that class of address. The first ioctl did not have a mask that matched the one in the primary address and would still cause the device address to be overwritten. The second ioctl that was called to set the mask would then detect as secondary and ignored, but the damage was already done. This was not an issue when using an application that used netlink sockets as the setting of IP and netmask came down at once. The inconsistent behavior between those two interfaces was something that needed to be resolved. While I was thinking about how I wanted to resolve this, Ralf Zeidler came with a patch that resolved this on a RHEL kernel by keeping a full shadow of the entries in dev->ifa_list for the bonding device and vlan devices in the bonding driver. I didn't like the duplication of the list as I want to see the 'bonding' struct and code shrink rather than grow, but liked the general idea. As the Subject indicates this patch drops the master_ip and vlan_ip elements from the 'bonding' and 'vlan_entry' structs, respectively. This can be done because a device's address-list is now traversed to determine the optimal source IP address for ARP requests and for checks to see if the bonding device has a particular IP address. This code could have all be contained inside the bonding driver, but it made more sense to me to EXPORT and call inet_confirm_addr since it did exactly what was needed. I tested this and a backported patch and everything works as expected. Ralf also helped with verification of the backported patch. Thanks to Ralf for all his help on this. v2: Whitespace and organizational changes based on suggestions from Jay Vosburgh and Dave Miller. v3: Fixup incorrect usage of rcu_read_unlock based on Dave Miller's suggestion. Signed-off-by: Andy Gospodarek <andy@greyhouse.net> CC: Ralf Zeidler <ralf.zeidler@nsn.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-22 20:14:29 +04:00
static inline __be32 bond_confirm_addr(struct net_device *dev, __be32 dst, __be32 local)
{
struct in_device *in_dev;
__be32 addr = 0;
rcu_read_lock();
in_dev = __in_dev_get_rcu(dev);
if (in_dev)
addr = inet_confirm_addr(dev_net(dev), in_dev, dst, local,
RT_SCOPE_HOST);
bonding: remove entries for master_ip and vlan_ip and query devices instead The following patch aimed to resolve an issue where secondary, tertiary, etc. addresses added to bond interfaces could overwrite the bond->master_ip and vlan_ip values. commit 917fbdb32f37e9a93b00bb12ee83532982982df3 Author: Henrik Saavedra Persson <henrik.e.persson@ericsson.com> Date: Wed Nov 23 23:37:15 2011 +0000 bonding: only use primary address for ARP That patch was good because it prevented bonds using ARP monitoring from sending frames with an invalid source IP address. Unfortunately, it didn't always work as expected. When using an ioctl (like ifconfig does) to set the IP address and netmask, 2 separate ioctls are actually called to set the IP and netmask if the mask chosen doesn't match the standard mask for that class of address. The first ioctl did not have a mask that matched the one in the primary address and would still cause the device address to be overwritten. The second ioctl that was called to set the mask would then detect as secondary and ignored, but the damage was already done. This was not an issue when using an application that used netlink sockets as the setting of IP and netmask came down at once. The inconsistent behavior between those two interfaces was something that needed to be resolved. While I was thinking about how I wanted to resolve this, Ralf Zeidler came with a patch that resolved this on a RHEL kernel by keeping a full shadow of the entries in dev->ifa_list for the bonding device and vlan devices in the bonding driver. I didn't like the duplication of the list as I want to see the 'bonding' struct and code shrink rather than grow, but liked the general idea. As the Subject indicates this patch drops the master_ip and vlan_ip elements from the 'bonding' and 'vlan_entry' structs, respectively. This can be done because a device's address-list is now traversed to determine the optimal source IP address for ARP requests and for checks to see if the bonding device has a particular IP address. This code could have all be contained inside the bonding driver, but it made more sense to me to EXPORT and call inet_confirm_addr since it did exactly what was needed. I tested this and a backported patch and everything works as expected. Ralf also helped with verification of the backported patch. Thanks to Ralf for all his help on this. v2: Whitespace and organizational changes based on suggestions from Jay Vosburgh and Dave Miller. v3: Fixup incorrect usage of rcu_read_unlock based on Dave Miller's suggestion. Signed-off-by: Andy Gospodarek <andy@greyhouse.net> CC: Ralf Zeidler <ralf.zeidler@nsn.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-22 20:14:29 +04:00
rcu_read_unlock();
return addr;
}
struct bond_net {
struct net *net; /* Associated network namespace */
struct list_head dev_list;
#ifdef CONFIG_PROC_FS
struct proc_dir_entry *proc_dir;
#endif
struct class_attribute class_attr_bonding_masters;
};
int bond_arp_rcv(const struct sk_buff *skb, struct bonding *bond, struct slave *slave);
void bond_dev_queue_xmit(struct bonding *bond, struct sk_buff *skb, struct net_device *slave_dev);
int bond_create(struct net *net, const char *name);
int bond_create_sysfs(struct bond_net *net);
void bond_destroy_sysfs(struct bond_net *net);
void bond_prepare_sysfs_group(struct bonding *bond);
int bond_sysfs_slave_add(struct slave *slave);
void bond_sysfs_slave_del(struct slave *slave);
int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev);
int bond_release(struct net_device *bond_dev, struct net_device *slave_dev);
u32 bond_xmit_hash(struct bonding *bond, struct sk_buff *skb);
int bond_set_carrier(struct bonding *bond);
void bond_select_active_slave(struct bonding *bond);
void bond_change_active_slave(struct bonding *bond, struct slave *new_active);
void bond_create_debugfs(void);
void bond_destroy_debugfs(void);
void bond_debug_register(struct bonding *bond);
void bond_debug_unregister(struct bonding *bond);
void bond_debug_reregister(struct bonding *bond);
const char *bond_mode_name(int mode);
void bond_setup(struct net_device *bond_dev);
unsigned int bond_get_num_tx_queues(void);
int bond_netlink_init(void);
void bond_netlink_fini(void);
struct net_device *bond_option_active_slave_get_rcu(struct bonding *bond);
const char *bond_slave_link_status(s8 link);
struct bond_vlan_tag *bond_verify_device_path(struct net_device *start_dev,
struct net_device *end_dev,
int level);
int bond_update_slave_arr(struct bonding *bond, struct slave *skipslave);
void bond_slave_arr_work_rearm(struct bonding *bond, unsigned long delay);
#ifdef CONFIG_PROC_FS
void bond_create_proc_entry(struct bonding *bond);
void bond_remove_proc_entry(struct bonding *bond);
void bond_create_proc_dir(struct bond_net *bn);
void bond_destroy_proc_dir(struct bond_net *bn);
#else
static inline void bond_create_proc_entry(struct bonding *bond)
{
}
static inline void bond_remove_proc_entry(struct bonding *bond)
{
}
static inline void bond_create_proc_dir(struct bond_net *bn)
{
}
static inline void bond_destroy_proc_dir(struct bond_net *bn)
{
}
#endif
static inline struct slave *bond_slave_has_mac(struct bonding *bond,
const u8 *mac)
{
struct list_head *iter;
struct slave *tmp;
bond_for_each_slave(bond, tmp, iter)
if (ether_addr_equal_64bits(mac, tmp->dev->dev_addr))
return tmp;
return NULL;
}
/* Caller must hold rcu_read_lock() for read */
static inline struct slave *bond_slave_has_mac_rcu(struct bonding *bond,
const u8 *mac)
{
struct list_head *iter;
struct slave *tmp;
bond_for_each_slave_rcu(bond, tmp, iter)
if (ether_addr_equal_64bits(mac, tmp->dev->dev_addr))
return tmp;
return NULL;
}
/* Caller must hold rcu_read_lock() for read */
static inline bool bond_slave_has_mac_rx(struct bonding *bond, const u8 *mac)
{
struct list_head *iter;
struct slave *tmp;
struct netdev_hw_addr *ha;
bond_for_each_slave_rcu(bond, tmp, iter)
if (ether_addr_equal_64bits(mac, tmp->dev->dev_addr))
return true;
if (netdev_uc_empty(bond->dev))
return false;
netdev_for_each_uc_addr(ha, bond->dev)
if (ether_addr_equal_64bits(mac, ha->addr))
return true;
return false;
}
/* Check if the ip is present in arp ip list, or first free slot if ip == 0
* Returns -1 if not found, index if found
*/
static inline int bond_get_targets_ip(__be32 *targets, __be32 ip)
{
int i;
for (i = 0; i < BOND_MAX_ARP_TARGETS; i++)
if (targets[i] == ip)
return i;
else if (targets[i] == 0)
break;
return -1;
}
/* exported from bond_main.c */
netns: make struct pernet_operations::id unsigned int Make struct pernet_operations::id unsigned. There are 2 reasons to do so: 1) This field is really an index into an zero based array and thus is unsigned entity. Using negative value is out-of-bound access by definition. 2) On x86_64 unsigned 32-bit data which are mixed with pointers via array indexing or offsets added or subtracted to pointers are preffered to signed 32-bit data. "int" being used as an array index needs to be sign-extended to 64-bit before being used. void f(long *p, int i) { g(p[i]); } roughly translates to movsx rsi, esi mov rdi, [rsi+...] call g MOVSX is 3 byte instruction which isn't necessary if the variable is unsigned because x86_64 is zero extending by default. Now, there is net_generic() function which, you guessed it right, uses "int" as an array index: static inline void *net_generic(const struct net *net, int id) { ... ptr = ng->ptr[id - 1]; ... } And this function is used a lot, so those sign extensions add up. Patch snipes ~1730 bytes on allyesconfig kernel (without all junk messing with code generation): add/remove: 0/0 grow/shrink: 70/598 up/down: 396/-2126 (-1730) Unfortunately some functions actually grow bigger. This is a semmingly random artefact of code generation with register allocator being used differently. gcc decides that some variable needs to live in new r8+ registers and every access now requires REX prefix. Or it is shifted into r12, so [r12+0] addressing mode has to be used which is longer than [r8] However, overall balance is in negative direction: add/remove: 0/0 grow/shrink: 70/598 up/down: 396/-2126 (-1730) function old new delta nfsd4_lock 3886 3959 +73 tipc_link_build_proto_msg 1096 1140 +44 mac80211_hwsim_new_radio 2776 2808 +32 tipc_mon_rcv 1032 1058 +26 svcauth_gss_legacy_init 1413 1429 +16 tipc_bcbase_select_primary 379 392 +13 nfsd4_exchange_id 1247 1260 +13 nfsd4_setclientid_confirm 782 793 +11 ... put_client_renew_locked 494 480 -14 ip_set_sockfn_get 730 716 -14 geneve_sock_add 829 813 -16 nfsd4_sequence_done 721 703 -18 nlmclnt_lookup_host 708 686 -22 nfsd4_lockt 1085 1063 -22 nfs_get_client 1077 1050 -27 tcf_bpf_init 1106 1076 -30 nfsd4_encode_fattr 5997 5930 -67 Total: Before=154856051, After=154854321, chg -0.00% Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-11-17 04:58:21 +03:00
extern unsigned int bond_net_id;
extern const struct bond_parm_tbl bond_lacp_tbl[];
extern const struct bond_parm_tbl xmit_hashtype_tbl[];
extern const struct bond_parm_tbl arp_validate_tbl[];
bonding: add an option to fail when any of arp_ip_target is inaccessible Currently, we fail only when all of the ips in arp_ip_target are gone. However, in some situations we might need to fail if even one host from arp_ip_target becomes unavailable. All situations, obviously, rely on the idea that we need *completely* functional network, with all interfaces/addresses working correctly. One real world example might be: vlans on top on bond (hybrid port). If bond and vlans have ips assigned and we have their peers monitored via arp_ip_target - in case of switch misconfiguration (trunk/access port), slave driver malfunction or tagged/untagged traffic dropped on the way - we will be able to switch to another slave. Though any other configuration needs that if we need to have access to all arp_ip_targets. This patch adds this possibility by adding a new parameter - arp_all_targets (both as a module parameter and as a sysfs knob). It can be set to: 0 or any (the default) - which works exactly as it's working now - the slave is up if any of the arp_ip_targets are up. 1 or all - the slave is up if all of the arp_ip_targets are up. This parameter can be changed on the fly (via sysfs), and requires the mode to be active-backup and arp_validate to be enabled (it obeys the arp_validate config on which slaves to validate). Internally it's done through: 1) Add target_last_arp_rx[BOND_MAX_ARP_TARGETS] array to slave struct. It's an array of jiffies, meaning that slave->target_last_arp_rx[i] is the last time we've received arp from bond->params.arp_targets[i] on this slave. 2) If we successfully validate an arp from bond->params.arp_targets[i] in bond_validate_arp() - update the slave->target_last_arp_rx[i] with the current jiffies value. 3) When getting slave's last_rx via slave_last_rx(), we return the oldest time when we've received an arp from any address in bond->params.arp_targets[]. If the value of arp_all_targets == 0 - we still work the same way as before. Also, update the documentation to reflect the new parameter. v3->v4: Kill the forgotten rtnl_unlock(), rephrase the documentation part to be more clear, don't fail setting arp_all_targets if arp_validate is not set - it has no effect anyway but can be easier to set up. Also, print a warning if the last arp_ip_target is removed while the arp_interval is on, but not the arp_validate. v2->v3: Use _bh spinlock, remove useless rtnl_lock() and use jiffies for new arp_ip_target last arp, instead of slave_last_rx(). On bond_enslave(), use the same initialization value for target_last_arp_rx[] as is used for the default last_arp_rx, to avoid useless interface flaps. Also, instead of failing to remove the last arp_ip_target just print a warning - otherwise it might break existing scripts. v1->v2: Correctly handle adding/removing hosts in arp_ip_target - we need to shift/initialize all slave's target_last_arp_rx. Also, don't fail module loading on arp_all_targets misconfiguration, just disable it, and some minor style fixes. Signed-off-by: Veaceslav Falico <vfalico@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-24 13:49:34 +04:00
extern const struct bond_parm_tbl arp_all_targets_tbl[];
extern const struct bond_parm_tbl fail_over_mac_tbl[];
extern const struct bond_parm_tbl pri_reselect_tbl[];
extern struct bond_parm_tbl ad_select_tbl[];
/* exported from bond_netlink.c */
extern struct rtnl_link_ops bond_link_ops;
static inline void bond_tx_drop(struct net_device *dev, struct sk_buff *skb)
{
atomic_long_inc(&dev->tx_dropped);
dev_kfree_skb_any(skb);
}
#endif /* _NET_BONDING_H */