[PATCH] bonding: rlb mode of bond should not alter ARP originatingvia bridge

From: zheng.li
Date: Tue Nov 27 2012 - 18:57:04 EST

Next message: Michal Hocko: "[PATCH] more debugging"
Previous message: Michal Hocko: "[PATCH] memcg: do not trigger OOM from add_to_page_cache_locked"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Do not modify or load balance ARP packets passing through balance-alb
mode (wherein the ARP did not originate locally, and arrived via a bridge).

Modifying pass-through ARP replies causes an incorrect MAC address
to be placed into the ARP packet, rendering peers unable to communicate
with the actual destination from which the ARP reply originated.

Load balancing pass-through ARP requests causes an entry to be
created for the peer in the rlb table, and bond_alb_monitor will
occasionally issue ARP updates to all peers in the table instrucing them
as to which MAC address they should communicate with; this occurs when
some event sets rx_ntt. In the bridged case, however, the MAC address
used for the update would be the MAC of the slave, not the actual source
MAC of the originating destination. This would render peers unable to
communicate with the destinations beyond the bridge.

Signed-off-by: Zheng Li <zheng.x.li@xxxxxxxxxx>
Cc: Jay Vosburgh <fubar@xxxxxxxxxx>
Cc: Andy Gospodarek <andy@xxxxxxxxxxxxx>
Cc: "David S. Miller" <davem@xxxxxxxxxxxxx>
Signed-off-by: Jay Vosburgh <fubar@xxxxxxxxxx>
Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>

Backports:
Signed-off-by: Matthew O'Connor (liquidhorse@xxxxxxxxx)

====== 3.0 Series ======

diff -uNr linux-3.0.0-a/drivers/net/bonding/bond_alb.c
linux-3.0.0-b/drivers/net/bonding/bond_alb.c
--- linux-3.0.0-a/drivers/net/bonding/bond_alb.c 2013-01-10
12:47:53.000000000 -0500
+++ linux-3.0.0-b/drivers/net/bonding/bond_alb.c 2013-01-10
12:50:58.000000000 -0500
@@ -666,6 +666,12 @@
struct arp_pkt *arp = arp_pkt(skb);
struct slave *tx_slave = NULL;

+ /* Don't modify or load balance ARPs that do not originate locally
+ * (e.g.,arrive via a bridge).
+ */
+ if (!bond_slave_has_mac(bond, arp->mac_src))
+ return NULL;
+
if (arp->op_code == htons(ARPOP_REPLY)) {
/* the arp must be sent on the selected
* rx channel
diff -uNr linux-3.0.0-a/drivers/net/bonding/bonding.h
linux-3.0.0-b/drivers/net/bonding/bonding.h
--- linux-3.0.0-a/drivers/net/bonding/bonding.h 2011-07-21
22:17:23.000000000 -0400
+++ linux-3.0.0-b/drivers/net/bonding/bonding.h 2013-01-10
12:51:05.000000000 -0500
@@ -18,6 +18,7 @@
#include <linux/timer.h>
#include <linux/proc_fs.h>
#include <linux/if_bonding.h>
+#include <linux/etherdevice.h>
#include <linux/cpumask.h>
#include <linux/in6.h>
#include <linux/netpoll.h>
@@ -431,6 +432,18 @@
}
#endif

+static inline struct slave *bond_slave_has_mac(struct bonding *bond,
+ const u8 *mac)
+{
+ int i = 0;
+ struct slave *tmp;
+
+ bond_for_each_slave(bond, tmp, i)
+ if (ether_addr_equal_64bits(mac, tmp->dev->dev_addr))
+ return tmp;
+
+ return NULL;
+}

/* exported from bond_main.c */
extern int bond_net_id;
diff -uNr linux-3.0.0-a/include/linux/etherdevice.h
linux-3.0.0-b/include/linux/etherdevice.h
--- linux-3.0.0-a/include/linux/etherdevice.h 2011-07-21
22:17:23.000000000 -0400
+++ linux-3.0.0-b/include/linux/etherdevice.h 2013-01-10
12:51:16.000000000 -0500
@@ -275,4 +275,37 @@
#endif
}

+/**
+ * ether_addr_equal_64bits - Compare two Ethernet addresses
+ * @addr1: Pointer to an array of 8 bytes
+ * @addr2: Pointer to an other array of 8 bytes
+ *
+ * Compare two Ethernet addresses, returns true if equal, false otherwise.
+ *
+ * The function doesn't need any conditional branches and possibly uses
+ * word memory accesses on CPU allowing cheap unaligned memory reads.
+ * arrays = { byte1, byte2, byte3, byte4, byte5, byte6, pad1, pad2 }
+ *
+ * Please note that alignment of addr1 & addr2 are only guaranteed to
be 16 bits.
+ */
+
+static inline bool ether_addr_equal_64bits(const u8 addr1[6+2],
+ const u8 addr2[6+2])
+{
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+ unsigned long fold = ((*(unsigned long *)addr1) ^
+ (*(unsigned long *)addr2));
+
+ if (sizeof(fold) == 8)
+ return zap_last_2bytes(fold) == 0;
+
+ fold |= zap_last_2bytes((*(unsigned long *)(addr1 + 4)) ^
+ (*(unsigned long *)(addr2 + 4)));
+ return fold == 0;
+#else
+ return ether_addr_equal(addr1, addr2);
+#endif
+}
+
+
#endif /* _LINUX_ETHERDEVICE_H */

====== 3.2 Series ======

diff -uprN linux-3.2.37/drivers/net/bonding/bond_alb.c
linux-3.2.37-patched/drivers/net/bonding/bond_alb.c
--- linux-3.2.37/drivers/net/bonding/bond_alb.c 2013-01-15
20:13:30.000000000 -0500
+++ linux-3.2.37-patched/drivers/net/bonding/bond_alb.c 2013-01-30
15:37:12.717485854 -0500
@@ -666,6 +666,12 @@ static struct slave *rlb_arp_xmit(struct
struct arp_pkt *arp = arp_pkt(skb);
struct slave *tx_slave = NULL;

+ /* Don't modify or load balance ARPs that do not originate locally
+ * (e.g.,arrive via a bridge).
+ */
+ if (!bond_slave_has_mac(bond, arp->mac_src))
+ return NULL;
+
if (arp->op_code == htons(ARPOP_REPLY)) {
/* the arp must be sent on the selected
* rx channel
diff -uprN linux-3.2.37/drivers/net/bonding/bonding.h
linux-3.2.37-patched/drivers/net/bonding/bonding.h
--- linux-3.2.37/drivers/net/bonding/bonding.h 2013-01-15
20:13:30.000000000 -0500
+++ linux-3.2.37-patched/drivers/net/bonding/bonding.h 2013-01-30
15:37:12.729485790 -0500
@@ -18,6 +18,7 @@
#include <linux/timer.h>
#include <linux/proc_fs.h>
#include <linux/if_bonding.h>
+#include <linux/etherdevice.h>
#include <linux/cpumask.h>
#include <linux/in6.h>
#include <linux/netpoll.h>
@@ -436,6 +437,18 @@ static inline void bond_destroy_proc_dir
}
#endif

+static inline struct slave *bond_slave_has_mac(struct bonding *bond,
+ const u8 *mac)
+{
+ int i = 0;
+ struct slave *tmp;
+
+ bond_for_each_slave(bond, tmp, i)
+ if (ether_addr_equal_64bits(mac, tmp->dev->dev_addr))
+ return tmp;
+
+ return NULL;
+}

/* exported from bond_main.c */
extern int bond_net_id;
diff -uprN linux-3.2.37/include/linux/etherdevice.h
linux-3.2.37-patched/include/linux/etherdevice.h
--- linux-3.2.37/include/linux/etherdevice.h 2013-01-15
20:13:30.000000000 -0500
+++ linux-3.2.37-patched/include/linux/etherdevice.h 2013-01-30
15:37:12.729485790 -0500
@@ -275,4 +275,37 @@ static inline unsigned long compare_ethe
#endif
}

+/**
+ * ether_addr_equal_64bits - Compare two Ethernet addresses
+ * @addr1: Pointer to an array of 8 bytes
+ * @addr2: Pointer to an other array of 8 bytes
+ *
+ * Compare two Ethernet addresses, returns true if equal, false otherwise.
+ *
+ * The function doesn't need any conditional branches and possibly uses
+ * word memory accesses on CPU allowing cheap unaligned memory reads.
+ * arrays = { byte1, byte2, byte3, byte4, byte5, byte6, pad1, pad2 }
+ *
+ * Please note that alignment of addr1 & addr2 are only guaranteed to
be 16 bits.
+ */
+
+static inline bool ether_addr_equal_64bits(const u8 addr1[6+2],
+ const u8 addr2[6+2])
+{
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+ unsigned long fold = ((*(unsigned long *)addr1) ^
+ (*(unsigned long *)addr2));
+
+ if (sizeof(fold) == 8)
+ return zap_last_2bytes(fold) == 0;
+
+ fold |= zap_last_2bytes((*(unsigned long *)(addr1 + 4)) ^
+ (*(unsigned long *)(addr2 + 4)));
+ return fold == 0;
+#else
+ return ether_addr_equal(addr1, addr2);
+#endif
+}
+
+
#endif /* _LINUX_ETHERDEVICE_H */

====== 3.4 Series ======

diff -uprN linux-3.4.28/drivers/net/bonding/bond_alb.c
linux-3.4.28-patched/drivers/net/bonding/bond_alb.c
--- linux-3.4.28/drivers/net/bonding/bond_alb.c 2013-01-27
23:51:45.000000000 -0500
+++ linux-3.4.28-patched/drivers/net/bonding/bond_alb.c 2013-01-30
15:37:25.121708311 -0500
@@ -704,6 +704,12 @@ static struct slave *rlb_arp_xmit(struct
struct arp_pkt *arp = arp_pkt(skb);
struct slave *tx_slave = NULL;

+ /* Don't modify or load balance ARPs that do not originate locally
+ * (e.g.,arrive via a bridge).
+ */
+ if (!bond_slave_has_mac(bond, arp->mac_src))
+ return NULL;
+
if (arp->op_code == htons(ARPOP_REPLY)) {
/* the arp must be sent on the selected
* rx channel
diff -uprN linux-3.4.28/drivers/net/bonding/bonding.h
linux-3.4.28-patched/drivers/net/bonding/bonding.h
--- linux-3.4.28/drivers/net/bonding/bonding.h 2013-01-27
23:51:45.000000000 -0500
+++ linux-3.4.28-patched/drivers/net/bonding/bonding.h 2013-01-30
15:37:25.121708311 -0500
@@ -18,6 +18,7 @@
#include <linux/timer.h>
#include <linux/proc_fs.h>
#include <linux/if_bonding.h>
+#include <linux/etherdevice.h>
#include <linux/cpumask.h>
#include <linux/in6.h>
#include <linux/netpoll.h>
@@ -450,6 +451,18 @@ static inline void bond_destroy_proc_dir
}
#endif

+static inline struct slave *bond_slave_has_mac(struct bonding *bond,
+ const u8 *mac)
+{
+ int i = 0;
+ struct slave *tmp;
+
+ bond_for_each_slave(bond, tmp, i)
+ if (ether_addr_equal_64bits(mac, tmp->dev->dev_addr))
+ return tmp;
+
+ return NULL;
+}

/* exported from bond_main.c */
extern int bond_net_id;
diff -uprN linux-3.4.28/include/linux/etherdevice.h
linux-3.4.28-patched/include/linux/etherdevice.h
--- linux-3.4.28/include/linux/etherdevice.h 2013-01-27
23:51:45.000000000 -0500
+++ linux-3.4.28-patched/include/linux/etherdevice.h 2013-01-30
15:37:25.121708311 -0500
@@ -277,4 +277,37 @@ static inline unsigned long compare_ethe
#endif
}

+/**
+ * ether_addr_equal_64bits - Compare two Ethernet addresses
+ * @addr1: Pointer to an array of 8 bytes
+ * @addr2: Pointer to an other array of 8 bytes
+ *
+ * Compare two Ethernet addresses, returns true if equal, false otherwise.
+ *
+ * The function doesn't need any conditional branches and possibly uses
+ * word memory accesses on CPU allowing cheap unaligned memory reads.
+ * arrays = { byte1, byte2, byte3, byte4, byte5, byte6, pad1, pad2 }
+ *
+ * Please note that alignment of addr1 & addr2 are only guaranteed to
be 16 bits.
+ */
+
+static inline bool ether_addr_equal_64bits(const u8 addr1[6+2],
+ const u8 addr2[6+2])
+{
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+ unsigned long fold = ((*(unsigned long *)addr1) ^
+ (*(unsigned long *)addr2));
+
+ if (sizeof(fold) == 8)
+ return zap_last_2bytes(fold) == 0;
+
+ fold |= zap_last_2bytes((*(unsigned long *)(addr1 + 4)) ^
+ (*(unsigned long *)(addr2 + 4)));
+ return fold == 0;
+#else
+ return ether_addr_equal(addr1, addr2);
+#endif
+}
+
+
#endif /* _LINUX_ETHERDEVICE_H */

====== 3.7 Series ======

diff -uprN linux-3.7/drivers/net/bonding/bond_alb.c
linux-3.7-patch2/drivers/net/bonding/bond_alb.c
--- linux-3.7/drivers/net/bonding/bond_alb.c 2012-12-10
22:30:57.000000000 -0500
+++ linux-3.7-patch2/drivers/net/bonding/bond_alb.c 2013-02-01
13:55:20.062030691 -0500
@@ -694,6 +694,12 @@ static struct slave *rlb_arp_xmit(struct
struct arp_pkt *arp = arp_pkt(skb);
struct slave *tx_slave = NULL;

+ /* Don't modify or load balance ARPs that do not originate locally
+ * (e.g.,arrive via a bridge).
+ */
+ if (!bond_slave_has_mac(bond, arp->mac_src))
+ return NULL;
+
if (arp->op_code == htons(ARPOP_REPLY)) {
/* the arp must be sent on the selected
* rx channel
diff -uprN linux-3.7/drivers/net/bonding/bonding.h
linux-3.7-patch2/drivers/net/bonding/bonding.h
--- linux-3.7/drivers/net/bonding/bonding.h 2012-12-10
22:30:57.000000000 -0500
+++ linux-3.7-patch2/drivers/net/bonding/bonding.h 2013-02-01
13:55:20.062030691 -0500
@@ -18,6 +18,7 @@
#include <linux/timer.h>
#include <linux/proc_fs.h>
#include <linux/if_bonding.h>
+#include <linux/etherdevice.h>
#include <linux/cpumask.h>
#include <linux/in6.h>
#include <linux/netpoll.h>
@@ -450,6 +451,18 @@ static inline void bond_destroy_proc_dir
}
#endif

+static inline struct slave *bond_slave_has_mac(struct bonding *bond,
+ const u8 *mac)
+{
+ int i = 0;
+ struct slave *tmp;
+
+ bond_for_each_slave(bond, tmp, i)
+ if (ether_addr_equal_64bits(mac, tmp->dev->dev_addr))
+ return tmp;
+
+ return NULL;
+}

/* exported from bond_main.c */
extern int bond_net_id;
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Next message: Michal Hocko: "[PATCH] more debugging"
Previous message: Michal Hocko: "[PATCH] memcg: do not trigger OOM from add_to_page_cache_locked"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]