Skip to content

L
linux
Commit 61b6a04a authored by Jeff Garzik's avatar Jeff Garzik
Browse files
Options

Merge redhat.com:/spare/repo/linux-2.5 

into redhat.com:/spare/repo/netdev-2.6/bonding-1
parents 0010bc50 cd0b043f
Show whitespace changes
Inline Side-by-side
Showing with 3820 additions and 3797 deletions
Documentation/networking/bonding.txt
View file @ 61b6a04a
......@@ -363,25 +363,11 @@ mode
2. Base driver support for setting the hw address of a
device also when it is open. This is required so that there
will always be one slave in the team using the bond hw
address (the current_slave) while having a unique hw
address for each slave in the bond. If the current_slave
fails it's hw address is swapped with the new current_slave
address (the curr_active_slave) while having a unique hw
address for each slave in the bond. If the curr_active_slave
fails it's hw address is swapped with the new curr_active_slave
that was chosen.
multicast
Option specifying the mode of operation for multicast support.
Possible values are:
disabled or 0
Disabled (no multicast support)
active or 1
Enabled on active slave only, useful in active-backup mode
all or 2
Enabled on all slaves, this is the default
primary
A string (eth0, eth2, etc) to equate to a primary device. If this
......@@ -973,13 +959,12 @@ Donald Becker's Ethernet Drivers and diag programs may be found at :
You will also find a lot of information regarding Ethernet, NWay, MII, etc. at
www.scyld.com.
For new versions of the driver, patches for older kernels and the updated
userspace tools, take a look at Willy Tarreau's site :
Patches for 2.2 kernels are at Willy Tarreau's site :
- http://wtarreau.free.fr/pub/bonding/
- http://www-miaif.lip6.fr/willy/pub/bonding/
- http://www-miaif.lip6.fr/~tarreau/pub/bonding/
To get latest informations about Linux Kernel development, please consult
the Linux Kernel Mailing List Archives at :
http://boudicca.tux.org/hypermail/linux-kernel/latest/
http://www.ussg.iu.edu/hypermail/linux/kernel/
-- END --
Documentation/networking/ifenslave.c
View file @ 61b6a04a
......@@ -4,8 +4,6 @@
* This program controls the Linux implementation of running multiple
* network interfaces in parallel.
*
* Usage: ifenslave [-v] master-interface < slave-interface [metric <N>] > ...
*
* Author: Donald Becker <becker@cesdis.gsfc.nasa.gov>
* Copyright 1994-1996 Donald Becker
*
......@@ -90,24 +88,30 @@
* - For opt_c: slave should not be set to the master's setting
* while it is running. It was already set during enslave. To
* simplify things, it is now handeled separately.
*
* - 2003/09/24 - Shmulik Hen <shmulik.hen at intel dot com>
* - Code cleanup and style changes
* set version to 1.1.0
*/
#define APP_VERSION "1.0.12"
#define APP_RELDATE "June 30, 2003"
#define APP_VERSION "1.1.0"
#define APP_RELDATE "Septemer 24, 2003"
#define APP_NAME "ifenslave"
static char *version =
APP_NAME ".c:v" APP_VERSION " (" APP_RELDATE ") " "\nDonald Becker (becker@cesdis.gsfc.nasa.gov).\n"
"detach support added on 2000/10/02 by Willy Tarreau (willy at meta-x.org).\n"
"2.4 kernel support added on 2001/02/16 by Chad N. Tindel (ctindel at ieee dot org.\n";
APP_NAME ".c:v" APP_VERSION " (" APP_RELDATE ")\n"
"o Donald Becker (becker@cesdis.gsfc.nasa.gov).\n"
"o Detach support added on 2000/10/02 by Willy Tarreau (willy at meta-x.org).\n"
"o 2.4 kernel support added on 2001/02/16 by Chad N. Tindel\n"
" (ctindel at ieee dot org).\n";
static const char *usage_msg =
"Usage: ifenslave [-adfrvVh] <master-interface> < <slave-if> [metric <N>] > ...\n"
" ifenslave -c master-interface slave-if\n";
"Usage: ifenslave [-f] <master-if> <slave-if> [<slave-if>...]\n"
" ifenslave -d <master-if> <slave-if> [<slave-if>...]\n"
" ifenslave -c <master-if> <slave-if>\n"
" ifenslave --help\n";
static const char *howto_msg =
"Usage: ifenslave [-adfrvVh] <master-interface> < <slave-if> [metric <N>] > ...\n"
" ifenslave -c master-interface slave-if\n"
static const char *help_msg =
"\n"
" To create a bond device, simply follow these three steps :\n"
" - ensure that the required drivers are properly loaded :\n"
......@@ -115,18 +119,32 @@ static const char *howto_msg =
" - assign an IP address to the bond device :\n"
" # ifconfig bond0 <addr> netmask <mask> broadcast <bcast>\n"
" - attach all the interfaces you need to the bond device :\n"
" # ifenslave bond0 eth0 eth1 eth2\n"
" # ifenslave [{-f|--force}] bond0 eth0 [eth1 [eth2]...]\n"
" If bond0 didn't have a MAC address, it will take eth0's. Then, all\n"
" interfaces attached AFTER this assignment will get the same MAC addr.\n"
"\n"
" To detach a dead interface without setting the bond device down :\n"
" # ifenslave -d bond0 eth1\n"
" (except for ALB/TLB modes)\n"
"\n"
" To set the bond device down and automatically release all the slaves :\n"
" # ifconfig bond0 down\n"
"\n"
" To detach a dead interface without setting the bond device down :\n"
" # ifenslave {-d|--detach} bond0 eth0 [eth1 [eth2]...]\n"
"\n"
" To change active slave :\n"
" # ifenslave -c bond0 eth0\n"
" # ifenslave {-c|--change-active} bond0 eth0\n"
"\n"
" To show master interface info\n"
" # ifenslave bond0\n"
"\n"
" To show all interfaces info\n"
" # ifenslave {-a|--all-interfaces}\n"
"\n"
" To be more verbose\n"
" # ifenslave {-v|--verbose} ...\n"
"\n"
" # ifenslave {-u|--usage} Show usage\n"
" # ifenslave {-V|--version} Show version\n"
" # ifenslave {-h|--help} This message\n"
"\n";
#include <unistd.h>
......@@ -155,474 +173,330 @@ typedef __uint8_t u8; /* ditto */
struct option longopts[] = {
/* { name has_arg *flag val } */
{"all-interfaces", 0, 0, 'a'}, /* Show all interfaces. */
{"change-active", 0, 0, 'c'}, /* Change the active slave. */
{"detach", 0, 0, 'd'}, /* Detach a slave interface. */
{"force", 0, 0, 'f'}, /* Force the operation. */
{"help", 0, 0, '?'}, /* Give help */
{"howto", 0, 0, 'h'}, /* Give some more help */
{"receive-slave", 0, 0, 'r'}, /* Make a receive-only slave. */
{"help", 0, 0, 'h'}, /* Give help */
{"usage", 0, 0, 'u'}, /* Give usage */
{"verbose", 0, 0, 'v'}, /* Report each action taken. */
{"version", 0, 0, 'V'}, /* Emit version information. */
{"detach", 0, 0, 'd'}, /* Detach a slave interface. */
{"change-active", 0, 0, 'c'}, /* Change the active slave. */
{ 0, 0, 0, 0 }
{ 0, 0, 0, 0}
};
/* Command-line flags. */
unsigned int
opt_a = 0, /* Show-all-interfaces flag. */
opt_c = 0, /* Change-active-slave flag. */
opt_d = 0, /* Detach a slave interface. */
opt_f = 0, /* Force the operation. */
opt_r = 0, /* Set up a Rx-only slave. */
opt_d = 0, /* detach a slave interface. */
opt_c = 0, /* change-active-slave flag. */
verbose = 0, /* Verbose flag. */
opt_version = 0,
opt_howto = 0;
int skfd = -1; /* AF_INET socket for ioctl() calls. */
opt_h = 0, /* Help */
opt_u = 0, /* Usage */
opt_v = 0, /* Verbose flag. */
opt_V = 0; /* Version */
int skfd = -1; /* AF_INET socket for ioctl() calls.*/
int abi_ver = 0; /* userland - kernel ABI version */
int hwaddr_set = 0; /* Master's hwaddr is set */
int saved_errno;
struct ifreq master_mtu, master_flags, master_hwaddr;
struct ifreq slave_mtu, slave_flags, slave_hwaddr;
struct dev_ifr {
struct ifreq *req_ifr;
char *req_name;
int req_type;
};
static void if_print(char *ifname);
static int get_abi_ver(char *master_ifname);
struct dev_ifr master_ifra[] = {
{&master_mtu, "SIOCGIFMTU", SIOCGIFMTU},
{&master_flags, "SIOCGIFFLAGS", SIOCGIFFLAGS},
{&master_hwaddr, "SIOCGIFHWADDR", SIOCGIFHWADDR},
{NULL, "", 0}
};
int
main(int argc, char **argv)
struct dev_ifr slave_ifra[] = {
{&slave_mtu, "SIOCGIFMTU", SIOCGIFMTU},
{&slave_flags, "SIOCGIFFLAGS", SIOCGIFFLAGS},
{&slave_hwaddr, "SIOCGIFHWADDR", SIOCGIFHWADDR},
{NULL, "", 0}
};
static void if_print(char *ifname);
static int get_drv_info(char *master_ifname);
static int get_if_settings(char *ifname, struct dev_ifr ifra[]);
static int get_slave_flags(char *slave_ifname);
static int set_master_hwaddr(char *master_ifname, struct sockaddr *hwaddr);
static int set_slave_hwaddr(char *slave_ifname, struct sockaddr *hwaddr);
static int set_slave_mtu(char *slave_ifname, int mtu);
static int set_if_flags(char *ifname, short flags);
static int set_if_up(char *ifname, short flags);
static int set_if_down(char *ifname, short flags);
static int clear_if_addr(char *ifname);
static int set_if_addr(char *master_ifname, char *slave_ifname);
static int change_active(char *master_ifname, char *slave_ifname);
static int enslave(char *master_ifname, char *slave_ifname);
static int release(char *master_ifname, char *slave_ifname);
#define v_print(fmt, args...) \
if (opt_v) \
fprintf(stderr, fmt, ## args )
int main(int argc, char *argv[])
{
struct ifreq ifr2, if_hwaddr, if_ipaddr, if_metric, if_mtu, if_dstaddr;
struct ifreq if_netmask, if_brdaddr, if_flags;
int rv, goterr = 0;
int c, errflag = 0;
sa_family_t master_family;
char **spp, *master_ifname, *slave_ifname;
int hwaddr_notset;
int abi_ver = 0;
int c, i, rv;
int res = 0;
int exclusive = 0;
while ((c = getopt_long(argc, argv, "acdfrvV?h", longopts, 0)) != EOF)
while ((c = getopt_long(argc, argv, "acdfhuvV", longopts, 0)) != EOF) {
switch (c) {
case 'a': opt_a++; break;
case 'f': opt_f++; break;
case 'r': opt_r++; break;
case 'd': opt_d++; break;
case 'c': opt_c++; break;
case 'v': verbose++; break;
case 'V': opt_version++; break;
case 'h': opt_howto++; break;
case '?': errflag++;
}
/* option check */
if (opt_c)
if(opt_a || opt_f || opt_r || opt_d || verbose || opt_version ||
opt_howto || errflag ) {
case 'a': opt_a++; exclusive++; break;
case 'c': opt_c++; exclusive++; break;
case 'd': opt_d++; exclusive++; break;
case 'f': opt_f++; exclusive++; break;
case 'h': opt_h++; exclusive++; break;
case 'u': opt_u++; exclusive++; break;
case 'v': opt_v++; break;
case 'V': opt_V++; exclusive++; break;
case '?':
fprintf(stderr, usage_msg);
return 2;
res = 2;
goto out;
}
}
if (errflag) {
/* options check */
if (exclusive > 1) {
fprintf(stderr, usage_msg);
return 2;
res = 2;
goto out;
}
if (opt_howto) {
fprintf(stderr, howto_msg);
return 0;
if (opt_v || opt_V) {
printf(version);
if (opt_V) {
res = 0;
goto out;
}
}
if (verbose || opt_version) {
printf(version);
if (opt_version)
exit(0);
if (opt_u) {
printf(usage_msg);
res = 0;
goto out;
}
/* Open a basic socket. */
if ((skfd = socket(AF_INET, SOCK_DGRAM,0)) < 0) {
perror("socket");
exit(-1);
if (opt_h) {
printf(usage_msg);
printf(help_msg);
res = 0;
goto out;
}
if (verbose)
fprintf(stderr, "DEBUG: argc=%d, optind=%d and argv[optind] is %s.\n",
argc, optind, argv[optind]);
/* Open a basic socket */
if ((skfd = socket(AF_INET, SOCK_DGRAM, 0)) < 0) {
perror("socket");
res = 1;
goto out;
}
/* No remaining args means show all interfaces. */
if (opt_a) {
if (optind == argc) {
/* No remaining args */
/* show all interfaces */
if_print((char *)NULL);
(void) close(skfd);
exit(0);
goto out;
} else {
/* Just show usage */
fprintf(stderr, usage_msg);
res = 2;
goto out;
}
}
/* Copy the interface name. */
/* Copy the interface name */
spp = argv + optind;
master_ifname = *spp++;
slave_ifname = *spp++;
/* Check command line. */
if (opt_c) {
char **tempp = spp;
if ((master_ifname == NULL)||(slave_ifname == NULL)||(*tempp++ != NULL)) {
if (master_ifname == NULL) {
fprintf(stderr, usage_msg);
(void) close(skfd);
return 2;
res = 2;
goto out;
}
}
/* A single args means show the configuration for this interface. */
if (slave_ifname == NULL) {
if_print(master_ifname);
(void) close(skfd);
exit(0);
}
/* exchange abi version with bonding driver */
abi_ver = get_abi_ver(master_ifname);
if (abi_ver < 0) {
(void) close(skfd);
exit(1);
}
/* Get the vitals from the master interface. */
{
struct ifreq *ifra[7] = { &if_ipaddr, &if_mtu, &if_dstaddr,
&if_brdaddr, &if_netmask, &if_flags,
&if_hwaddr };
const char *req_name[7] = {
"IP address", "MTU", "destination address",
"broadcast address", "netmask", "status flags",
"hardware address" };
const int ioctl_req_type[7] = {
SIOCGIFADDR, SIOCGIFMTU, SIOCGIFDSTADDR,
SIOCGIFBRDADDR, SIOCGIFNETMASK, SIOCGIFFLAGS,
SIOCGIFHWADDR };
int i;
for (i = 0; i < 7; i++) {
strncpy(ifra[i]->ifr_name, master_ifname, IFNAMSIZ);
if (ioctl(skfd, ioctl_req_type[i], ifra[i]) < 0) {
/* exchange abi version with bonding module */
res = get_drv_info(master_ifname);
if (res) {
fprintf(stderr,
"Something broke getting the master's %s: %s.\n",
req_name[i], strerror(errno));
}
}
/* check if master is up; if not then fail any operation */
if (!(if_flags.ifr_flags & IFF_UP)) {
fprintf(stderr, "Illegal operation; the specified master interface '%s' is not up.\n", master_ifname);
(void) close(skfd);
exit (1);
}
hwaddr_notset = 1; /* assume master's address not set yet */
for (i = 0; hwaddr_notset && (i < 6); i++) {
hwaddr_notset &= ((unsigned char *)if_hwaddr.ifr_hwaddr.sa_data)[i] == 0;
}
/* The family '1' is ARPHRD_ETHER for ethernet. */
if (if_hwaddr.ifr_hwaddr.sa_family != 1 && !opt_f) {
fprintf(stderr, "The specified master interface '%s' is not"
" ethernet-like.\n This program is designed to work"
" with ethernet-like network interfaces.\n"
" Use the '-f' option to force the operation.\n",
"Master '%s': Error: handshake with driver failed. "
"Aborting\n",
master_ifname);
(void) close(skfd);
exit (1);
}
master_family = if_hwaddr.ifr_hwaddr.sa_family;
if (verbose) {
unsigned char *hwaddr = (unsigned char *)if_hwaddr.ifr_hwaddr.sa_data;
printf("The current hardware address (SIOCGIFHWADDR) of %s is type %d "
"%2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x.\n", master_ifname,
if_hwaddr.ifr_hwaddr.sa_family, hwaddr[0], hwaddr[1],
hwaddr[2], hwaddr[3], hwaddr[4], hwaddr[5]);
}
goto out;
}
slave_ifname = *spp++;
/* do this when enslaving interfaces */
do {
if (opt_d) { /* detach a slave interface from the master */
strncpy(if_flags.ifr_name, master_ifname, IFNAMSIZ);
strncpy(if_flags.ifr_slave, slave_ifname, IFNAMSIZ);
if ((ioctl(skfd, SIOCBONDRELEASE, &if_flags) < 0) &&
(ioctl(skfd, BOND_RELEASE_OLD, &if_flags) < 0)) {
fprintf(stderr, "SIOCBONDRELEASE: cannot detach %s from %s. errno=%s.\n",
slave_ifname, master_ifname, strerror(errno));
}
else if (abi_ver < 1) {
/* The driver is using an old ABI, so we'll set the interface
* down to avoid any conflicts due to same IP/MAC
*/
strncpy(ifr2.ifr_name, slave_ifname, IFNAMSIZ);
if (ioctl(skfd, SIOCGIFFLAGS, &ifr2) < 0) {
int saved_errno = errno;
fprintf(stderr, "SIOCGIFFLAGS on %s failed: %s\n", slave_ifname,
strerror(saved_errno));
}
else {
ifr2.ifr_flags &= ~(IFF_UP | IFF_RUNNING);
if (ioctl(skfd, SIOCSIFFLAGS, &ifr2) < 0) {
int saved_errno = errno;
fprintf(stderr, "Shutting down interface %s failed: %s\n",
slave_ifname, strerror(saved_errno));
}
}
}
} else if (opt_c) { /* change primary slave */
strncpy(if_flags.ifr_name, master_ifname, IFNAMSIZ);
strncpy(if_flags.ifr_slave, slave_ifname, IFNAMSIZ);
if ((ioctl(skfd, SIOCBONDCHANGEACTIVE, &if_flags) < 0) &&
(ioctl(skfd, BOND_CHANGE_ACTIVE_OLD, &if_flags) < 0)) {
fprintf(stderr, "SIOCBONDCHANGEACTIVE: %s.\n", strerror(errno));
if (slave_ifname == NULL) {
if (opt_d || opt_c) {
fprintf(stderr, usage_msg);
res = 2;
goto out;
}
} else { /* attach a slave interface to the master */
strncpy(ifr2.ifr_name, slave_ifname, IFNAMSIZ);
if (ioctl(skfd, SIOCGIFFLAGS, &ifr2) < 0) {
int saved_errno = errno;
fprintf(stderr, "SIOCGIFFLAGS on %s failed: %s\n", slave_ifname,
strerror(saved_errno));
(void) close(skfd);
return 1;
/* A single arg means show the
* configuration for this interface
*/
if_print(master_ifname);
goto out;
}
if ((ifr2.ifr_flags & IFF_SLAVE) && !opt_r) {
fprintf(stderr, "%s is already a slave\n", slave_ifname);
(void) close(skfd);
return 1;
res = get_if_settings(master_ifname, master_ifra);
if (res) {
/* Probably a good reason not to go on */
fprintf(stderr,
"Master '%s': Error: get settings failed: %s. "
"Aborting\n",
master_ifname, strerror(res));
goto out;
}
/* if hwaddr_notset, assign the slave hw address to the master */
if (hwaddr_notset) {
/* assign the slave hw address to the
* master since it currently does not
* have one; otherwise, slaves may
* have different hw addresses in
* active-backup mode as seen when enslaving
* using "ifenslave bond0 eth0 eth1" because
* hwaddr_notset is set outside this loop.
* TODO: put this and the "else" portion in
* a function.
/* check if master is indeed a master;
* if not then fail any operation
*/
/* get the slaves MAC address */
strncpy(if_hwaddr.ifr_name, slave_ifname,
IFNAMSIZ);
rv = ioctl(skfd, SIOCGIFHWADDR, &if_hwaddr);
if (-1 == rv) {
fprintf(stderr, "Could not get MAC "
"address of %s: %s\n",
slave_ifname,
strerror(errno));
strncpy(if_hwaddr.ifr_name,
master_ifname, IFNAMSIZ);
goterr = 1;
}
if (!goterr) {
if (abi_ver < 1) {
/* In ABI versions older than 1, the
* master's set_mac routine couldn't
* work if it was up, because it
* used the default ethernet set_mac
* function.
*/
/* bring master down */
if_flags.ifr_flags &= ~IFF_UP;
if (ioctl(skfd, SIOCSIFFLAGS,
&if_flags) < 0) {
goterr = 1;
if (!(master_flags.ifr_flags & IFF_MASTER)) {
fprintf(stderr,
"Shutting down "
"interface %s failed: "
"%s\n",
master_ifname,
strerror(errno));
}
"Illegal operation; the specified interface '%s' "
"is not a master. Aborting\n",
master_ifname);
res = 1;
goto out;
}
strncpy(if_hwaddr.ifr_name,
master_ifname, IFNAMSIZ);
if (ioctl(skfd, SIOCSIFHWADDR,
&if_hwaddr) < 0) {
/* check if master is up; if not then fail any operation */
if (!(master_flags.ifr_flags & IFF_UP)) {
fprintf(stderr,
"Could not set MAC "
"address of %s: %s\n",
master_ifname,
strerror(errno));
goterr=1;
} else {
hwaddr_notset = 0;
"Illegal operation; the specified master interface "
"'%s' is not up.\n",
master_ifname);
res = 1;
goto out;
}
if (abi_ver < 1) {
/* bring master back up */
if_flags.ifr_flags |= IFF_UP;
if (ioctl(skfd, SIOCSIFFLAGS,
&if_flags) < 0) {
fprintf(stderr,
"Bringing up interface "
"%s failed: %s\n",
master_ifname,
strerror(errno));
}
}
}
} else if (abi_ver < 1) { /* if (hwaddr_notset) */
/* Only for enslaving */
if (!opt_c && !opt_d) {
sa_family_t master_family = master_hwaddr.ifr_hwaddr.sa_family;
unsigned char *hwaddr =
(unsigned char *)master_hwaddr.ifr_hwaddr.sa_data;
/* The driver is using an old ABI, so we'll set the interface
* down and assign the master's hwaddr to it
*/
if (ifr2.ifr_flags & IFF_UP) {
ifr2.ifr_flags &= ~IFF_UP;
if (ioctl(skfd, SIOCSIFFLAGS, &ifr2) < 0) {
int saved_errno = errno;
fprintf(stderr, "Shutting down interface %s failed: %s\n",
slave_ifname, strerror(saved_errno));
}
/* The family '1' is ARPHRD_ETHER for ethernet. */
if (master_family != 1 && !opt_f) {
fprintf(stderr,
"Illegal operation: The specified master "
"interface '%s' is not ethernet-like.\n "
"This program is designed to work with "
"ethernet-like network interfaces.\n "
"Use the '-f' option to force the "
"operation.\n",
master_ifname);
res = 1;
goto out;
}
strncpy(if_hwaddr.ifr_name, slave_ifname, IFNAMSIZ);
if (ioctl(skfd, SIOCSIFHWADDR, &if_hwaddr) < 0) {
int saved_errno = errno;
fprintf(stderr, "SIOCSIFHWADDR on %s failed: %s\n", if_hwaddr.ifr_name,
strerror(saved_errno));
if (saved_errno == EBUSY)
fprintf(stderr, " The slave device %s is busy: it must be"
" idle before running this command.\n", slave_ifname);
else if (saved_errno == EOPNOTSUPP)
fprintf(stderr, " The slave device you specified does not support"
" setting the MAC address.\n Your kernel likely does not"
" support slave devices.\n");
else if (saved_errno == EINVAL)
fprintf(stderr, " The slave device's address type does not match"
" the master's address type.\n");
} else {
if (verbose) {
unsigned char *hwaddr = if_hwaddr.ifr_hwaddr.sa_data;
printf("Slave's (%s) hardware address set to "
"%2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x.\n", slave_ifname,
hwaddr[0], hwaddr[1], hwaddr[2], hwaddr[3], hwaddr[4], hwaddr[5]);
}
/* Check master's hw addr */
for (i = 0; i < 6; i++) {
if (hwaddr[i] != 0) {
hwaddr_set = 1;
break;
}
}
if (*spp && !strcmp(*spp, "metric")) {
if (*++spp == NULL) {
fprintf(stderr, usage_msg);
(void) close(skfd);
exit(2);
}
if_metric.ifr_metric = atoi(*spp);
strncpy(if_metric.ifr_name, slave_ifname, IFNAMSIZ);
if (ioctl(skfd, SIOCSIFMETRIC, &if_metric) < 0) {
fprintf(stderr, "SIOCSIFMETRIC on %s: %s\n", slave_ifname,
strerror(errno));
goterr = 1;
if (hwaddr_set) {
v_print("current hardware address of master '%s' "
"is %2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x, "
"type %d\n",
master_ifname,
hwaddr[0], hwaddr[1],
hwaddr[2], hwaddr[3],
hwaddr[4], hwaddr[5],
master_family);
}
spp++;
}
if (strncpy(if_ipaddr.ifr_name, slave_ifname, IFNAMSIZ) <= 0
|| ioctl(skfd, SIOCSIFADDR, &if_ipaddr) < 0) {
/* Accepts only one slave */
if (opt_c) {
/* change active slave */
res = get_slave_flags(slave_ifname);
if (res) {
fprintf(stderr,
"Something broke setting the slave's address: %s.\n",
strerror(errno));
} else {
if (verbose) {
unsigned char *ipaddr = if_ipaddr.ifr_addr.sa_data;
printf("Set the slave's (%s) IP address to %d.%d.%d.%d.\n",
slave_ifname, ipaddr[0], ipaddr[1], ipaddr[2], ipaddr[3]);
"Slave '%s': Error: get flags failed. "
"Aborting\n",
slave_ifname);
goto out;
}
res = change_active(master_ifname, slave_ifname);
if (res) {
fprintf(stderr,
"Master '%s', Slave '%s': Error: "
"Change active failed\n",
master_ifname, slave_ifname);
}
if (strncpy(if_mtu.ifr_name, slave_ifname, IFNAMSIZ) <= 0
|| ioctl(skfd, SIOCSIFMTU, &if_mtu) < 0) {
fprintf(stderr, "Something broke setting the slave MTU: %s.\n",
strerror(errno));
} else {
if (verbose)
printf("Set the slave's (%s) MTU to %d.\n", slave_ifname, if_mtu.ifr_mtu);
}
if (strncpy(if_dstaddr.ifr_name, slave_ifname, IFNAMSIZ) <= 0
|| ioctl(skfd, SIOCSIFDSTADDR, &if_dstaddr) < 0) {
fprintf(stderr, "Error setting the slave (%s) with SIOCSIFDSTADDR: %s.\n",
slave_ifname, strerror(errno));
} else {
if (verbose) {
unsigned char *ipaddr = if_dstaddr.ifr_dstaddr.sa_data;
printf("Set the slave's (%s) destination address to %d.%d.%d.%d.\n",
slave_ifname, ipaddr[0], ipaddr[1], ipaddr[2], ipaddr[3]);
}
}
if (strncpy(if_brdaddr.ifr_name, slave_ifname, IFNAMSIZ) <= 0
|| ioctl(skfd, SIOCSIFBRDADDR, &if_brdaddr) < 0) {
/* Accept multiple slaves */
do {
if (opt_d) {
/* detach a slave interface from the master */
rv = get_slave_flags(slave_ifname);
if (rv) {
/* Can't work with this slave. */
/* remember the error and skip it*/
fprintf(stderr,
"Something broke setting the slave (%s) broadcast address: %s.\n",
slave_ifname, strerror(errno));
} else {
if (verbose) {
unsigned char *ipaddr = if_brdaddr.ifr_broadaddr.sa_data;
printf("Set the slave's (%s) broadcast address to %d.%d.%d.%d.\n",
slave_ifname, ipaddr[0], ipaddr[1], ipaddr[2], ipaddr[3]);
}
"Slave '%s': Error: get flags "
"failed. Skipping\n",
slave_ifname);
res = rv;
continue;
}
if (strncpy(if_netmask.ifr_name, slave_ifname, IFNAMSIZ) <= 0
|| ioctl(skfd, SIOCSIFNETMASK, &if_netmask) < 0) {
rv = release(master_ifname, slave_ifname);
if (rv) {
fprintf(stderr,
"Something broke setting the slave (%s) netmask: %s.\n",
slave_ifname, strerror(errno));
} else {
if (verbose) {
unsigned char *ipaddr = if_netmask.ifr_netmask.sa_data;
printf("Set the slave's (%s) netmask to %d.%d.%d.%d.\n",
slave_ifname, ipaddr[0], ipaddr[1], ipaddr[2], ipaddr[3]);
}
"Master '%s', Slave '%s': Error: "
"Release failed\n",
master_ifname, slave_ifname);
res = rv;
}
if (abi_ver < 1) {
/* The driver is using an old ABI, so we'll set the interface
* up before enslaving it
*/
ifr2.ifr_flags |= IFF_UP;
if ((ifr2.ifr_flags &= ~(IFF_SLAVE | IFF_MASTER)) == 0
|| strncpy(ifr2.ifr_name, slave_ifname, IFNAMSIZ) <= 0
|| ioctl(skfd, SIOCSIFFLAGS, &ifr2) < 0) {
fprintf(stderr,
"Something broke setting the slave (%s) flags: %s.\n",
slave_ifname, strerror(errno));
} else {
if (verbose)
printf("Set the slave's (%s) flags %4.4x.\n",
slave_ifname, if_flags.ifr_flags);
/* attach a slave interface to the master */
rv = get_if_settings(slave_ifname, slave_ifra);
if (rv) {
/* Can't work with this slave. */
/* remember the error and skip it*/
fprintf(stderr,
"Slave '%s': Error: get "
"settings failed: %s. "
"Skipping\n",
slave_ifname, strerror(rv));
res = rv;
continue;
}
} else {
/* the bonding module takes care of setting the slave's mac address
* and opening its interface
*/
if (ifr2.ifr_flags & IFF_UP) { /* the interface will need to be down */
ifr2.ifr_flags &= ~IFF_UP;
if (ioctl(skfd, SIOCSIFFLAGS, &ifr2) < 0) {
int saved_errno = errno;
fprintf(stderr, "Shutting down interface %s failed: %s\n",
slave_ifname, strerror(saved_errno));
rv = enslave(master_ifname, slave_ifname);
if (rv) {
fprintf(stderr,
"Master '%s', Slave '%s': Error: "
"Enslave failed\n",
master_ifname, slave_ifname);
res = rv;
}
}
} while ((slave_ifname = *spp++) != NULL);
}
/* Do the real thing */
if (!opt_r) {
strncpy(if_flags.ifr_name, master_ifname, IFNAMSIZ);
strncpy(if_flags.ifr_slave, slave_ifname, IFNAMSIZ);
if ((ioctl(skfd, SIOCBONDENSLAVE, &if_flags) < 0) &&
(ioctl(skfd, BOND_ENSLAVE_OLD, &if_flags) < 0)) {
fprintf(stderr, "SIOCBONDENSLAVE: %s.\n", strerror(errno));
}
}
out:
if (skfd >= 0) {
close(skfd);
}
} while ( (slave_ifname = *spp++) != NULL);
/* Close the socket. */
(void) close(skfd);
return(goterr);
return res;
}
static short mif_flags;
......@@ -633,6 +507,7 @@ static int if_getconfig(char *ifname)
struct ifreq ifr;
int metric, mtu; /* Parameters of the master interface. */
struct sockaddr dstaddr, broadaddr, netmask;
unsigned char *hwaddr;
strcpy(ifr.ifr_name, ifname);
if (ioctl(skfd, SIOCGIFFLAGS, &ifr) < 0)
......@@ -652,14 +527,12 @@ static int if_getconfig(char *ifname)
if (ioctl(skfd, SIOCGIFHWADDR, &ifr) < 0)
return -1;
{
/* Gotta convert from 'char' to unsigned for printf(). */
unsigned char *hwaddr = (unsigned char *)ifr.ifr_hwaddr.sa_data;
hwaddr = (unsigned char *)ifr.ifr_hwaddr.sa_data;
printf("The result of SIOCGIFHWADDR is type %d "
"%2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x.\n",
ifr.ifr_hwaddr.sa_family, hwaddr[0], hwaddr[1],
hwaddr[2], hwaddr[3], hwaddr[4], hwaddr[5]);
}
strcpy(ifr.ifr_name, ifname);
if (ioctl(skfd, SIOCGIFMETRIC, &ifr) < 0) {
......@@ -691,7 +564,7 @@ static int if_getconfig(char *ifname)
} else
netmask = ifr.ifr_netmask;
return(0);
return 0;
}
static void if_print(char *ifname)
......@@ -705,14 +578,15 @@ static void if_print(char *ifname)
ifc.ifc_len = sizeof(buff);
ifc.ifc_buf = buff;
if (ioctl(skfd, SIOCGIFCONF, &ifc) < 0) {
fprintf(stderr, "SIOCGIFCONF: %s\n", strerror(errno));
perror("SIOCGIFCONF failed");
return;
}
ifr = ifc.ifc_req;
for (i = ifc.ifc_len / sizeof(struct ifreq); --i >= 0; ifr++) {
if (if_getconfig(ifr->ifr_name) < 0) {
fprintf(stderr, "%s: unknown interface.\n",
fprintf(stderr,
"%s: unknown interface.\n",
ifr->ifr_name);
continue;
}
......@@ -721,16 +595,18 @@ static void if_print(char *ifname)
/*ife_print(&ife);*/
}
} else {
if (if_getconfig(ifname) < 0)
fprintf(stderr, "%s: unknown interface.\n", ifname);
if (if_getconfig(ifname) < 0) {
fprintf(stderr,
"%s: unknown interface.\n", ifname);
}
}
}
static int get_abi_ver(char *master_ifname)
static int get_drv_info(char *master_ifname)
{
struct ifreq ifr;
struct ethtool_drvinfo info;
int abi_ver = 0;
char *endptr;
memset(&ifr, 0, sizeof(ifr));
strncpy(ifr.ifr_name, master_ifname, IFNAMSIZ);
......@@ -739,24 +615,487 @@ static int get_abi_ver(char *master_ifname)
info.cmd = ETHTOOL_GDRVINFO;
strncpy(info.driver, "ifenslave", 32);
snprintf(info.fw_version, 32, "%d", BOND_ABI_VERSION);
if (ioctl(skfd, SIOCETHTOOL, &ifr) >= 0) {
char *endptr;
if (ioctl(skfd, SIOCETHTOOL, &ifr) < 0) {
if (errno == EOPNOTSUPP) {
goto out;
}
saved_errno = errno;
v_print("Master '%s': Error: get bonding info failed %s\n",
master_ifname, strerror(saved_errno));
return 1;
}
abi_ver = strtoul(info.fw_version, &endptr, 0);
if (*endptr) {
fprintf(stderr, "Error: got invalid string as an ABI "
"version from the bonding module\n");
return -1;
v_print("Master '%s': Error: got invalid string as an ABI "
"version from the bonding module\n",
master_ifname);
return 1;
}
out:
v_print("ABI ver is %d\n", abi_ver);
return 0;
}
static int change_active(char *master_ifname, char *slave_ifname)
{
struct ifreq ifr;
int res = 0;
if (!(slave_flags.ifr_flags & IFF_SLAVE)) {
fprintf(stderr,
"Illegal operation: The specified slave interface "
"'%s' is not a slave\n",
slave_ifname);
return 1;
}
strncpy(ifr.ifr_name, master_ifname, IFNAMSIZ);
strncpy(ifr.ifr_slave, slave_ifname, IFNAMSIZ);
if ((ioctl(skfd, SIOCBONDCHANGEACTIVE, &ifr) < 0) &&
(ioctl(skfd, BOND_CHANGE_ACTIVE_OLD, &ifr) < 0)) {
saved_errno = errno;
v_print("Master '%s': Error: SIOCBONDCHANGEACTIVE failed: "
"%s\n",
master_ifname, strerror(saved_errno));
res = 1;
}
return res;
}
static int enslave(char *master_ifname, char *slave_ifname)
{
struct ifreq ifr;
int res = 0;
if (slave_flags.ifr_flags & IFF_SLAVE) {
fprintf(stderr,
"Illegal operation: The specified slave interface "
"'%s' is already a slave\n",
slave_ifname);
return 1;
}
res = set_if_down(slave_ifname, slave_flags.ifr_flags);
if (res) {
fprintf(stderr,
"Slave '%s': Error: bring interface down failed\n",
slave_ifname);
return res;
}
if (abi_ver < 2) {
/* Older bonding versions would panic if the slave has no IP
* address, so get the IP setting from the master.
*/
res = set_if_addr(master_ifname, slave_ifname);
if (res) {
fprintf(stderr,
"Slave '%s': Error: set address failed\n",
slave_ifname);
return res;
}
} else {
res = clear_if_addr(slave_ifname);
if (res) {
fprintf(stderr,
"Slave '%s': Error: clear address failed\n",
slave_ifname);
return res;
}
}
if (master_mtu.ifr_mtu != slave_mtu.ifr_mtu) {
res = set_slave_mtu(slave_ifname, master_mtu.ifr_mtu);
if (res) {
fprintf(stderr,
"Slave '%s': Error: set MTU failed\n",
slave_ifname);
return res;
}
}
if (hwaddr_set) {
/* Master already has an hwaddr
* so set it's hwaddr to the slave
*/
if (abi_ver < 1) {
/* The driver is using an old ABI, so
* the application sets the slave's
* hwaddr
*/
res = set_slave_hwaddr(slave_ifname,
&(master_hwaddr.ifr_hwaddr));
if (res) {
fprintf(stderr,
"Slave '%s': Error: set hw address "
"failed\n",
slave_ifname);
goto undo_mtu;
}
/* For old ABI the application needs to bring the
* slave back up
*/
res = set_if_up(slave_ifname, slave_flags.ifr_flags);
if (res) {
fprintf(stderr,
"Slave '%s': Error: bring interface "
"down failed\n",
slave_ifname);
goto undo_slave_mac;
}
}
/* The driver is using a new ABI,
* so the driver takes care of setting
* the slave's hwaddr and bringing
* it up again
*/
} else {
/* No hwaddr for master yet, so
* set the slave's hwaddr to it
*/
if (abi_ver < 1) {
/* For old ABI, the master needs to be
* down before setting it's hwaddr
*/
res = set_if_down(master_ifname, master_flags.ifr_flags);
if (res) {
fprintf(stderr,
"Master '%s': Error: bring interface "
"down failed\n",
master_ifname);
goto undo_mtu;
}
}
res = set_master_hwaddr(master_ifname,
&(slave_hwaddr.ifr_hwaddr));
if (res) {
fprintf(stderr,
"Master '%s': Error: set hw address "
"failed\n",
master_ifname);
goto undo_mtu;
}
if (abi_ver < 1) {
/* For old ABI, bring the master
* back up
*/
res = set_if_up(master_ifname, master_flags.ifr_flags);
if (res) {
fprintf(stderr,
"Master '%s': Error: bring interface "
"up failed\n",
master_ifname);
goto undo_master_mac;
}
}
hwaddr_set = 1;
}
/* Do the real thing */
strncpy(ifr.ifr_name, master_ifname, IFNAMSIZ);
strncpy(ifr.ifr_slave, slave_ifname, IFNAMSIZ);
if ((ioctl(skfd, SIOCBONDENSLAVE, &ifr) < 0) &&
(ioctl(skfd, BOND_ENSLAVE_OLD, &ifr) < 0)) {
saved_errno = errno;
v_print("Master '%s': Error: SIOCBONDENSLAVE failed: %s\n",
master_ifname, strerror(saved_errno));
res = 1;
}
if (res) {
goto undo_master_mac;
}
return 0;
/* rollback (best effort) */
undo_master_mac:
set_master_hwaddr(master_ifname, &(master_hwaddr.ifr_hwaddr));
hwaddr_set = 0;
goto undo_mtu;
undo_slave_mac:
set_slave_hwaddr(slave_ifname, &(slave_hwaddr.ifr_hwaddr));
undo_mtu:
set_slave_mtu(slave_ifname, slave_mtu.ifr_mtu);
return res;
}
static int release(char *master_ifname, char *slave_ifname)
{
struct ifreq ifr;
int res = 0;
if (!(slave_flags.ifr_flags & IFF_SLAVE)) {
fprintf(stderr,
"Illegal operation: The specified slave interface "
"'%s' is not a slave\n",
slave_ifname);
return 1;
}
strncpy(ifr.ifr_name, master_ifname, IFNAMSIZ);
strncpy(ifr.ifr_slave, slave_ifname, IFNAMSIZ);
if ((ioctl(skfd, SIOCBONDRELEASE, &ifr) < 0) &&
(ioctl(skfd, BOND_RELEASE_OLD, &ifr) < 0)) {
saved_errno = errno;
v_print("Master '%s': Error: SIOCBONDRELEASE failed: %s\n",
master_ifname, strerror(saved_errno));
return 1;
} else if (abi_ver < 1) {
/* The driver is using an old ABI, so we'll set the interface
* down to avoid any conflicts due to same MAC/IP
*/
res = set_if_down(slave_ifname, slave_flags.ifr_flags);
if (res) {
fprintf(stderr,
"Slave '%s': Error: bring interface "
"down failed\n",
slave_ifname);
}
}
/* set to default mtu */
set_slave_mtu(slave_ifname, 1500);
return res;
}
static int get_if_settings(char *ifname, struct dev_ifr ifra[])
{
int i;
int res = 0;
for (i = 0; ifra[i].req_ifr; i++) {
strncpy(ifra[i].req_ifr->ifr_name, ifname, IFNAMSIZ);
res = ioctl(skfd, ifra[i].req_type, ifra[i].req_ifr);
if (res < 0) {
saved_errno = errno;
v_print("Interface '%s': Error: %s failed: %s\n",
ifname, ifra[i].req_name,
strerror(saved_errno));
return saved_errno;
}
}
if (verbose) {
printf("ABI ver is %d\n", abi_ver);
return 0;
}
static int get_slave_flags(char *slave_ifname)
{
int res = 0;
strncpy(slave_flags.ifr_name, slave_ifname, IFNAMSIZ);
res = ioctl(skfd, SIOCGIFFLAGS, &slave_flags);
if (res < 0) {
saved_errno = errno;
v_print("Slave '%s': Error: SIOCGIFFLAGS failed: %s\n",
slave_ifname, strerror(saved_errno));
} else {
v_print("Slave %s: flags %04X.\n",
slave_ifname, slave_flags.ifr_flags);
}
return res;
}
static int set_master_hwaddr(char *master_ifname, struct sockaddr *hwaddr)
{
unsigned char *addr = (unsigned char *)hwaddr->sa_data;
struct ifreq ifr;
int res = 0;
strncpy(ifr.ifr_name, master_ifname, IFNAMSIZ);
memcpy(&(ifr.ifr_hwaddr), hwaddr, sizeof(struct sockaddr));
res = ioctl(skfd, SIOCSIFHWADDR, &ifr);
if (res < 0) {
saved_errno = errno;
v_print("Master '%s': Error: SIOCSIFHWADDR failed: %s\n",
master_ifname, strerror(saved_errno));
return res;
} else {
v_print("Master '%s': hardware address set to "
"%2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x.\n",
master_ifname, addr[0], addr[1], addr[2],
addr[3], addr[4], addr[5]);
}
return abi_ver;
return res;
}
static int set_slave_hwaddr(char *slave_ifname, struct sockaddr *hwaddr)
{
unsigned char *addr = (unsigned char *)hwaddr->sa_data;
struct ifreq ifr;
int res = 0;
strncpy(ifr.ifr_name, slave_ifname, IFNAMSIZ);
memcpy(&(ifr.ifr_hwaddr), hwaddr, sizeof(struct sockaddr));
res = ioctl(skfd, SIOCSIFHWADDR, &ifr);
if (res < 0) {
saved_errno = errno;
v_print("Slave '%s': Error: SIOCSIFHWADDR failed: %s\n",
slave_ifname, strerror(saved_errno));
if (saved_errno == EBUSY) {
v_print(" The device is busy: it must be idle "
"before running this command.\n");
} else if (saved_errno == EOPNOTSUPP) {
v_print(" The device does not support setting "
"the MAC address.\n"
" Your kernel likely does not support slave "
"devices.\n");
} else if (saved_errno == EINVAL) {
v_print(" The device's address type does not match "
"the master's address type.\n");
}
return res;
} else {
v_print("Slave '%s': hardware address set to "
"%2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x.\n",
slave_ifname, addr[0], addr[1], addr[2],
addr[3], addr[4], addr[5]);
}
return res;
}
static int set_slave_mtu(char *slave_ifname, int mtu)
{
struct ifreq ifr;
int res = 0;
ifr.ifr_mtu = mtu;
strncpy(ifr.ifr_name, slave_ifname, IFNAMSIZ);
res = ioctl(skfd, SIOCSIFMTU, &ifr);
if (res < 0) {
saved_errno = errno;
v_print("Slave '%s': Error: SIOCSIFMTU failed: %s\n",
slave_ifname, strerror(saved_errno));
} else {
v_print("Slave '%s': MTU set to %d.\n", slave_ifname, mtu);
}
return res;
}
static int set_if_flags(char *ifname, short flags)
{
struct ifreq ifr;
int res = 0;
ifr.ifr_flags = flags;
strncpy(ifr.ifr_name, ifname, IFNAMSIZ);
res = ioctl(skfd, SIOCSIFFLAGS, &ifr);
if (res < 0) {
saved_errno = errno;
v_print("Interface '%s': Error: SIOCSIFFLAGS failed: %s\n",
ifname, strerror(saved_errno));
} else {
v_print("Interface '%s': flags set to %04X.\n", ifname, flags);
}
return res;
}
static int set_if_up(char *ifname, short flags)
{
return set_if_flags(ifname, flags | IFF_UP);
}
static int set_if_down(char *ifname, short flags)
{
return set_if_flags(ifname, flags & ~IFF_UP);
}
static int clear_if_addr(char *ifname)
{
struct ifreq ifr;
int res = 0;
strncpy(ifr.ifr_name, ifname, IFNAMSIZ);
ifr.ifr_addr.sa_family = AF_INET;
memset(ifr.ifr_addr.sa_data, 0, sizeof(ifr.ifr_addr.sa_data));
res = ioctl(skfd, SIOCSIFADDR, &ifr);
if (res < 0) {
saved_errno = errno;
v_print("Interface '%s': Error: SIOCSIFADDR failed: %s\n",
ifname, strerror(saved_errno));
} else {
v_print("Interface '%s': address cleared\n", ifname);
}
return res;
}
static int set_if_addr(char *master_ifname, char *slave_ifname)
{
struct ifreq ifr;
int res;
unsigned char *ipaddr;
int i;
struct {
char *req_name;
char *desc;
int g_ioctl;
int s_ioctl;
} ifra[] = {
{"IFADDR", "addr", SIOCGIFADDR, SIOCSIFADDR},
{"DSTADDR", "destination addr", SIOCGIFDSTADDR, SIOCSIFDSTADDR},
{"BRDADDR", "broadcast addr", SIOCGIFBRDADDR, SIOCSIFBRDADDR},
{"NETMASK", "netmask", SIOCGIFNETMASK, SIOCSIFNETMASK},
{NULL, NULL, 0, 0},
};
for (i = 0; ifra[i].req_name; i++) {
strncpy(ifr.ifr_name, master_ifname, IFNAMSIZ);
res = ioctl(skfd, ifra[i].g_ioctl, &ifr);
if (res < 0) {
int saved_errno = errno;
v_print("Interface '%s': Error: SIOCG%s failed: %s\n",
master_ifname, ifra[i].req_name,
strerror(saved_errno));
ifr.ifr_addr.sa_family = AF_INET;
memset(ifr.ifr_addr.sa_data, 0,
sizeof(ifr.ifr_addr.sa_data));
}
strncpy(ifr.ifr_name, slave_ifname, IFNAMSIZ);
res = ioctl(skfd, ifra[i].s_ioctl, &ifr);
if (res < 0) {
int saved_errno = errno;
v_print("Interface '%s': Error: SIOCS%s failed: %s\n",
slave_ifname, ifra[i].req_name,
strerror(saved_errno));
return res;
}
ipaddr = ifr.ifr_addr.sa_data;
v_print("Interface '%s': set IP %s to %d.%d.%d.%d\n",
slave_ifname, ifra[i].desc,
ipaddr[0], ipaddr[1], ipaddr[2], ipaddr[3]);
}
return 0;
}
/*
* Local variables:
......@@ -768,3 +1107,4 @@ static int get_abi_ver(char *master_ifname)
* compile-command: "gcc -Wall -Wstrict-prototypes -O -I/usr/src/linux/include ifenslave.c -o ifenslave"
* End:
*/
drivers/net/bonding/bond_3ad.c
View file @ 61b6a04a
......@@ -47,8 +47,13 @@
* - Send LACPDU as highest priority packet to further fix the above
* problem on very high Tx traffic load where packets may get dropped
* by the slave.
*
* 2003/09/24 - Shmulik Hen <shmulik.hen at intel dot com>
* - Code cleanup and style changes
*/
//#define BONDING_DEBUG 1
#include <linux/skbuff.h>
#include <linux/if_ether.h>
#include <linux/netdevice.h>
......@@ -119,6 +124,7 @@
static struct mac_addr null_mac_addr = {{0, 0, 0, 0, 0, 0}};
static u16 ad_ticks_per_sec;
static const int ad_delta_in_ticks = (AD_TIMER_INTERVAL * HZ) / 1000;
// ================= 3AD api to bonding and kernel code ==================
static u16 __get_link_speed(struct port *port);
......@@ -196,13 +202,11 @@ static inline struct bonding *__get_bond_by_port(struct port *port)
*/
static inline struct port *__get_first_port(struct bonding *bond)
{
struct slave *slave = bond->next;
if (slave == (struct slave *)bond) {
if (bond->slave_cnt == 0) {
return NULL;
}
return &(SLAVE_AD_INFO(slave).port);
return &(SLAVE_AD_INFO(bond->first_slave).port);
}
/**
......@@ -218,7 +222,7 @@ static inline struct port *__get_next_port(struct port *port)
struct slave *slave = port->slave;
// If there's no bond for this port, or this is the last slave
if ((bond == NULL) || (slave->next == bond->next)) {
if ((bond == NULL) || (slave->next == bond->first_slave)) {
return NULL;
}
......@@ -236,12 +240,12 @@ static inline struct aggregator *__get_first_agg(struct port *port)
{
struct bonding *bond = __get_bond_by_port(port);
// If there's no bond for this port, or this is the last slave
if ((bond == NULL) || (bond->next == (struct slave *)bond)) {
// If there's no bond for this port, or bond has no slaves
if ((bond == NULL) || (bond->slave_cnt == 0)) {
return NULL;
}
return &(SLAVE_AD_INFO(bond->next).aggregator);
return &(SLAVE_AD_INFO(bond->first_slave).aggregator);
}
/**
......@@ -257,7 +261,7 @@ static inline struct aggregator *__get_next_agg(struct aggregator *aggregator)
struct bonding *bond = bond_get_bond_by_slave(slave);
// If there's no bond for this aggregator, or this is the last slave
if ((bond == NULL) || (slave->next == bond->next)) {
if ((bond == NULL) || (slave->next == bond->first_slave)) {
return NULL;
}
......@@ -392,7 +396,7 @@ static u16 __get_link_speed(struct port *port)
}
}
BOND_PRINT_DBG(("Port %d Received link speed %d update from adapter", port->actor_port_number, speed));
dprintk("Port %d Received link speed %d update from adapter\n", port->actor_port_number, speed);
return speed;
}
......@@ -418,12 +422,12 @@ static u8 __get_duplex(struct port *port)
switch (slave->duplex) {
case DUPLEX_FULL:
retval=0x1;
BOND_PRINT_DBG(("Port %d Received status full duplex update from adapter", port->actor_port_number));
dprintk("Port %d Received status full duplex update from adapter\n", port->actor_port_number);
break;
case DUPLEX_HALF:
default:
retval=0x0;
BOND_PRINT_DBG(("Port %d Received status NOT full duplex update from adapter", port->actor_port_number));
dprintk("Port %d Received status NOT full duplex update from adapter\n", port->actor_port_number);
break;
}
}
......@@ -1059,7 +1063,7 @@ static void ad_mux_machine(struct port *port)
// check if the state machine was changed
if (port->sm_mux_state != last_state) {
BOND_PRINT_DBG(("Mux Machine: Port=%d, Last State=%d, Curr State=%d", port->actor_port_number, last_state, port->sm_mux_state));
dprintk("Mux Machine: Port=%d, Last State=%d, Curr State=%d\n", port->actor_port_number, last_state, port->sm_mux_state);
switch (port->sm_mux_state) {
case AD_MUX_DETACHED:
__detach_bond_from_agg(port);
......@@ -1158,7 +1162,7 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
// check if the State machine was changed or new lacpdu arrived
if ((port->sm_rx_state != last_state) || (lacpdu)) {
BOND_PRINT_DBG(("Rx Machine: Port=%d, Last State=%d, Curr State=%d", port->actor_port_number, last_state, port->sm_rx_state));
dprintk("Rx Machine: Port=%d, Last State=%d, Curr State=%d\n", port->actor_port_number, last_state, port->sm_rx_state);
switch (port->sm_rx_state) {
case AD_RX_INITIALIZE:
if (!(port->actor_oper_port_key & AD_DUPLEX_KEY_BITS)) {
......@@ -1204,7 +1208,7 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
// detect loopback situation
if (!MAC_ADDRESS_COMPARE(&(lacpdu->actor_system), &(port->actor_system))) {
// INFO_RECEIVED_LOOPBACK_FRAMES
printk(KERN_ERR "bonding: An illegal loopback occurred on adapter (%s)\n",
printk(KERN_ERR DRV_NAME ": An illegal loopback occurred on adapter (%s)\n",
port->slave->dev->name);
printk(KERN_ERR "Check the configuration to verify that all Adapters "
"are connected to 802.3ad compliant switch ports\n");
......@@ -1245,7 +1249,7 @@ static void ad_tx_machine(struct port *port)
__update_lacpdu_from_port(port);
// send the lacpdu
if (ad_lacpdu_send(port) >= 0) {
BOND_PRINT_DBG(("Sent LACPDU on port %d", port->actor_port_number));
dprintk("Sent LACPDU on port %d\n", port->actor_port_number);
// mark ntt as false, so it will not be sent again until demanded
port->ntt = 0;
}
......@@ -1318,7 +1322,7 @@ static void ad_periodic_machine(struct port *port)
// check if the state machine was changed
if (port->sm_periodic_state != last_state) {
BOND_PRINT_DBG(("Periodic Machine: Port=%d, Last State=%d, Curr State=%d", port->actor_port_number, last_state, port->sm_periodic_state));
dprintk("Periodic Machine: Port=%d, Last State=%d, Curr State=%d\n", port->actor_port_number, last_state, port->sm_periodic_state);
switch (port->sm_periodic_state) {
case AD_NO_PERIODIC:
port->sm_periodic_timer_counter = 0; // zero timer
......@@ -1375,7 +1379,7 @@ static void ad_port_selection_logic(struct port *port)
port->next_port_in_aggregator=NULL;
port->actor_port_aggregator_identifier=0;
BOND_PRINT_DBG(("Port %d left LAG %d", port->actor_port_number, temp_aggregator->aggregator_identifier));
dprintk("Port %d left LAG %d\n", port->actor_port_number, temp_aggregator->aggregator_identifier);
// if the aggregator is empty, clear its parameters, and set it ready to be attached
if (!temp_aggregator->lag_ports) {
ad_clear_agg(temp_aggregator);
......@@ -1384,7 +1388,7 @@ static void ad_port_selection_logic(struct port *port)
}
}
if (!curr_port) { // meaning: the port was related to an aggregator but was not on the aggregator port list
printk(KERN_WARNING "bonding: Warning: Port %d (on %s) was "
printk(KERN_WARNING DRV_NAME ": Warning: Port %d (on %s) was "
"related to aggregator %d but was not on its port list\n",
port->actor_port_number, port->slave->dev->name,
port->aggregator->aggregator_identifier);
......@@ -1417,7 +1421,7 @@ static void ad_port_selection_logic(struct port *port)
port->next_port_in_aggregator=aggregator->lag_ports;
port->aggregator->num_of_ports++;
aggregator->lag_ports=port;
BOND_PRINT_DBG(("Port %d joined LAG %d(existing LAG)", port->actor_port_number, port->aggregator->aggregator_identifier));
dprintk("Port %d joined LAG %d(existing LAG)\n", port->actor_port_number, port->aggregator->aggregator_identifier);
// mark this port as selected
port->sm_vars |= AD_PORT_SELECTED;
......@@ -1454,9 +1458,9 @@ static void ad_port_selection_logic(struct port *port)
// mark this port as selected
port->sm_vars |= AD_PORT_SELECTED;
BOND_PRINT_DBG(("Port %d joined LAG %d(new LAG)", port->actor_port_number, port->aggregator->aggregator_identifier));
dprintk("Port %d joined LAG %d(new LAG)\n", port->actor_port_number, port->aggregator->aggregator_identifier);
} else {
printk(KERN_ERR "bonding: Port %d (on %s) did not find a suitable aggregator\n",
printk(KERN_ERR DRV_NAME ": Port %d (on %s) did not find a suitable aggregator\n",
port->actor_port_number, port->slave->dev->name);
}
}
......@@ -1580,30 +1584,30 @@ static void ad_agg_selection_logic(struct aggregator *aggregator)
aggregator;
aggregator = __get_next_agg(aggregator)) {
BOND_PRINT_DBG(("Agg=%d; Ports=%d; a key=%d; p key=%d; Indiv=%d; Active=%d",
dprintk("Agg=%d; Ports=%d; a key=%d; p key=%d; Indiv=%d; Active=%d\n",
aggregator->aggregator_identifier, aggregator->num_of_ports,
aggregator->actor_oper_aggregator_key, aggregator->partner_oper_aggregator_key,
aggregator->is_individual, aggregator->is_active));
aggregator->is_individual, aggregator->is_active);
}
// check if any partner replys
if (best_aggregator->is_individual) {
printk(KERN_WARNING "bonding: Warning: No 802.3ad response from the link partner "
printk(KERN_WARNING DRV_NAME ": Warning: No 802.3ad response from the link partner "
"for any adapters in the bond\n");
}
// check if there are more than one aggregator
if (num_of_aggs > 1) {
BOND_PRINT_DBG(("Warning: More than one Link Aggregation Group was "
"found in the bond. Only one group will function in the bond"));
dprintk("Warning: More than one Link Aggregation Group was "
"found in the bond. Only one group will function in the bond\n");
}
best_aggregator->is_active = 1;
BOND_PRINT_DBG(("LAG %d choosed as the active LAG", best_aggregator->aggregator_identifier));
BOND_PRINT_DBG(("Agg=%d; Ports=%d; a key=%d; p key=%d; Indiv=%d; Active=%d",
dprintk("LAG %d choosed as the active LAG\n", best_aggregator->aggregator_identifier);
dprintk("Agg=%d; Ports=%d; a key=%d; p key=%d; Indiv=%d; Active=%d\n",
best_aggregator->aggregator_identifier, best_aggregator->num_of_ports,
best_aggregator->actor_oper_aggregator_key, best_aggregator->partner_oper_aggregator_key,
best_aggregator->is_individual, best_aggregator->is_active));
best_aggregator->is_individual, best_aggregator->is_active);
// disable the ports that were related to the former active_aggregator
if (last_active_aggregator) {
......@@ -1644,7 +1648,7 @@ static void ad_clear_agg(struct aggregator *aggregator)
aggregator->lag_ports = NULL;
aggregator->is_active = 0;
aggregator->num_of_ports = 0;
BOND_PRINT_DBG(("LAG %d was cleared", aggregator->aggregator_identifier));
dprintk("LAG %d was cleared\n", aggregator->aggregator_identifier);
}
}
......@@ -1729,7 +1733,7 @@ static void ad_initialize_port(struct port *port, int lacp_fast)
static void ad_enable_collecting_distributing(struct port *port)
{
if (port->aggregator->is_active) {
BOND_PRINT_DBG(("Enabling port %d(LAG %d)", port->actor_port_number, port->aggregator->aggregator_identifier));
dprintk("Enabling port %d(LAG %d)\n", port->actor_port_number, port->aggregator->aggregator_identifier);
__enable_port(port);
}
}
......@@ -1742,7 +1746,7 @@ static void ad_enable_collecting_distributing(struct port *port)
static void ad_disable_collecting_distributing(struct port *port)
{
if (port->aggregator && MAC_ADDRESS_COMPARE(&(port->aggregator->partner_system), &(null_mac_addr))) {
BOND_PRINT_DBG(("Disabling port %d(LAG %d)", port->actor_port_number, port->aggregator->aggregator_identifier));
dprintk("Disabling port %d(LAG %d)\n", port->actor_port_number, port->aggregator->aggregator_identifier);
__disable_port(port);
}
}
......@@ -1780,7 +1784,7 @@ static void ad_marker_info_send(struct port *port)
// send the marker information
if (ad_marker_send(port, &marker) >= 0) {
BOND_PRINT_DBG(("Sent Marker Information on port %d", port->actor_port_number));
dprintk("Sent Marker Information on port %d\n", port->actor_port_number);
}
}
#endif
......@@ -1803,7 +1807,7 @@ static void ad_marker_info_received(struct marker *marker_info,struct port *port
// send the marker response
if (ad_marker_send(port, &marker) >= 0) {
BOND_PRINT_DBG(("Sent Marker Response on port %d", port->actor_port_number));
dprintk("Sent Marker Response on port %d\n", port->actor_port_number);
}
}
......@@ -1890,13 +1894,13 @@ static u16 aggregator_identifier;
void bond_3ad_initialize(struct bonding *bond, u16 tick_resolution, int lacp_fast)
{
// check that the bond is not initialized yet
if (MAC_ADDRESS_COMPARE(&(BOND_AD_INFO(bond).system.sys_mac_addr), &(bond->device->dev_addr))) {
if (MAC_ADDRESS_COMPARE(&(BOND_AD_INFO(bond).system.sys_mac_addr), &(bond->dev->dev_addr))) {
aggregator_identifier = 0;
BOND_AD_INFO(bond).lacp_fast = lacp_fast;
BOND_AD_INFO(bond).system.sys_priority = 0xFFFF;
BOND_AD_INFO(bond).system.sys_mac_addr = *((struct mac_addr *)bond->device->dev_addr);
BOND_AD_INFO(bond).system.sys_mac_addr = *((struct mac_addr *)bond->dev->dev_addr);
// initialize how many times this module is called in one second(should be about every 100ms)
ad_ticks_per_sec = tick_resolution;
......@@ -1921,7 +1925,7 @@ int bond_3ad_bind_slave(struct slave *slave)
struct aggregator *aggregator;
if (bond == NULL) {
printk(KERN_CRIT "The slave %s is not attached to its bond\n", slave->dev->name);
printk(KERN_ERR "The slave %s is not attached to its bond\n", slave->dev->name);
return -1;
}
......@@ -1964,7 +1968,7 @@ int bond_3ad_bind_slave(struct slave *slave)
ad_initialize_agg(aggregator);
aggregator->aggregator_mac_address = *((struct mac_addr *)bond->device->dev_addr);
aggregator->aggregator_mac_address = *((struct mac_addr *)bond->dev->dev_addr);
aggregator->aggregator_identifier = (++aggregator_identifier);
aggregator->slave = slave;
aggregator->is_active = 0;
......@@ -1996,11 +2000,11 @@ void bond_3ad_unbind_slave(struct slave *slave)
// if slave is null, the whole port is not initialized
if (!port->slave) {
printk(KERN_WARNING "bonding: Trying to unbind an uninitialized port on %s\n", slave->dev->name);
printk(KERN_WARNING DRV_NAME ": Trying to unbind an uninitialized port on %s\n", slave->dev->name);
return;
}
BOND_PRINT_DBG(("Unbinding Link Aggregation Group %d", aggregator->aggregator_identifier));
dprintk("Unbinding Link Aggregation Group %d\n", aggregator->aggregator_identifier);
/* Tell the partner that this port is not suitable for aggregation */
port->actor_oper_port_state &= ~AD_STATE_AGGREGATION;
......@@ -2024,10 +2028,10 @@ void bond_3ad_unbind_slave(struct slave *slave)
// if new aggregator found, copy the aggregator's parameters
// and connect the related lag_ports to the new aggregator
if ((new_aggregator) && ((!new_aggregator->lag_ports) || ((new_aggregator->lag_ports == port) && !new_aggregator->lag_ports->next_port_in_aggregator))) {
BOND_PRINT_DBG(("Some port(s) related to LAG %d - replaceing with LAG %d", aggregator->aggregator_identifier, new_aggregator->aggregator_identifier));
dprintk("Some port(s) related to LAG %d - replaceing with LAG %d\n", aggregator->aggregator_identifier, new_aggregator->aggregator_identifier);
if ((new_aggregator->lag_ports == port) && new_aggregator->is_active) {
printk(KERN_INFO "bonding: Removing an active aggregator\n");
printk(KERN_INFO DRV_NAME ": Removing an active aggregator\n");
// select new active aggregator
select_new_active_agg = 1;
}
......@@ -2057,7 +2061,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
ad_agg_selection_logic(__get_first_agg(port));
}
} else {
printk(KERN_WARNING "bonding: Warning: unbinding aggregator, "
printk(KERN_WARNING DRV_NAME ": Warning: unbinding aggregator, "
"and could not find a new aggregator for its ports\n");
}
} else { // in case that the only port related to this aggregator is the one we want to remove
......@@ -2072,7 +2076,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
}
}
BOND_PRINT_DBG(("Unbinding port %d", port->actor_port_number));
dprintk("Unbinding port %d\n", port->actor_port_number);
// find the aggregator that this port is connected to
temp_aggregator = __get_first_agg(port);
for (; temp_aggregator; temp_aggregator = __get_next_agg(temp_aggregator)) {
......@@ -2123,13 +2127,13 @@ void bond_3ad_state_machine_handler(struct bonding *bond)
read_lock(&bond->lock);
//check if there are any slaves
if (bond->next == (struct slave *)bond) {
goto end;
if (bond->kill_timers) {
goto out;
}
if ((bond->device->flags & IFF_UP) != IFF_UP) {
goto end;
//check if there are any slaves
if (bond->slave_cnt == 0) {
goto re_arm;
}
// check if agg_select_timer timer after initialize is timed out
......@@ -2137,8 +2141,8 @@ void bond_3ad_state_machine_handler(struct bonding *bond)
// select the active aggregator for the bond
if ((port = __get_first_port(bond))) {
if (!port->slave) {
printk(KERN_WARNING "bonding: Warning: bond's first port is uninitialized\n");
goto end;
printk(KERN_WARNING DRV_NAME ": Warning: bond's first port is uninitialized\n");
goto re_arm;
}
aggregator = __get_first_agg(port);
......@@ -2149,8 +2153,8 @@ void bond_3ad_state_machine_handler(struct bonding *bond)
// for each port run the state machines
for (port = __get_first_port(bond); port; port = __get_next_port(port)) {
if (!port->slave) {
printk(KERN_WARNING "bonding: Warning: Found an uninitialized port\n");
goto end;
printk(KERN_WARNING DRV_NAME ": Warning: Found an uninitialized port\n");
goto re_arm;
}
ad_rx_machine(NULL, port);
......@@ -2165,14 +2169,10 @@ void bond_3ad_state_machine_handler(struct bonding *bond)
}
}
end:
re_arm:
mod_timer(&(BOND_AD_INFO(bond).ad_timer), jiffies + ad_delta_in_ticks);
out:
read_unlock(&bond->lock);
if ((bond->device->flags & IFF_UP) == IFF_UP) {
/* re-arm the timer */
mod_timer(&(BOND_AD_INFO(bond).ad_timer), jiffies + (AD_TIMER_INTERVAL * HZ / 1000));
}
}
/**
......@@ -2194,14 +2194,14 @@ void bond_3ad_rx_indication(struct lacpdu *lacpdu, struct slave *slave, u16 leng
port = &(SLAVE_AD_INFO(slave).port);
if (!port->slave) {
printk(KERN_WARNING "bonding: Warning: port of slave %s is uninitialized\n", slave->dev->name);
printk(KERN_WARNING DRV_NAME ": Warning: port of slave %s is uninitialized\n", slave->dev->name);
return;
}
switch (lacpdu->subtype) {
case AD_TYPE_LACPDU:
__ntohs_lacpdu(lacpdu);
BOND_PRINT_DBG(("Received LACPDU on port %d", port->actor_port_number));
dprintk("Received LACPDU on port %d\n", port->actor_port_number);
ad_rx_machine(lacpdu, port);
break;
......@@ -2210,17 +2210,17 @@ void bond_3ad_rx_indication(struct lacpdu *lacpdu, struct slave *slave, u16 leng
switch (((struct marker *)lacpdu)->tlv_type) {
case AD_MARKER_INFORMATION_SUBTYPE:
BOND_PRINT_DBG(("Received Marker Information on port %d", port->actor_port_number));
dprintk("Received Marker Information on port %d\n", port->actor_port_number);
ad_marker_info_received((struct marker *)lacpdu, port);
break;
case AD_MARKER_RESPONSE_SUBTYPE:
BOND_PRINT_DBG(("Received Marker Response on port %d", port->actor_port_number));
dprintk("Received Marker Response on port %d\n", port->actor_port_number);
ad_marker_response_received((struct marker *)lacpdu, port);
break;
default:
BOND_PRINT_DBG(("Received an unknown Marker subtype on slot %d", port->actor_port_number));
dprintk("Received an unknown Marker subtype on slot %d\n", port->actor_port_number);
}
}
}
......@@ -2240,14 +2240,14 @@ void bond_3ad_adapter_speed_changed(struct slave *slave)
// if slave is null, the whole port is not initialized
if (!port->slave) {
printk(KERN_WARNING "bonding: Warning: speed changed for uninitialized port on %s\n",
printk(KERN_WARNING DRV_NAME ": Warning: speed changed for uninitialized port on %s\n",
slave->dev->name);
return;
}
port->actor_admin_port_key &= ~AD_SPEED_KEY_BITS;
port->actor_oper_port_key=port->actor_admin_port_key |= (__get_link_speed(port) << 1);
BOND_PRINT_DBG(("Port %d changed speed", port->actor_port_number));
dprintk("Port %d changed speed\n", port->actor_port_number);
// there is no need to reselect a new aggregator, just signal the
// state machines to reinitialize
port->sm_vars |= AD_PORT_BEGIN;
......@@ -2267,14 +2267,14 @@ void bond_3ad_adapter_duplex_changed(struct slave *slave)
// if slave is null, the whole port is not initialized
if (!port->slave) {
printk(KERN_WARNING "bonding: Warning: duplex changed for uninitialized port on %s\n",
printk(KERN_WARNING DRV_NAME ": Warning: duplex changed for uninitialized port on %s\n",
slave->dev->name);
return;
}
port->actor_admin_port_key &= ~AD_DUPLEX_KEY_BITS;
port->actor_oper_port_key=port->actor_admin_port_key |= __get_duplex(port);
BOND_PRINT_DBG(("Port %d changed duplex", port->actor_port_number));
dprintk("Port %d changed duplex\n", port->actor_port_number);
// there is no need to reselect a new aggregator, just signal the
// state machines to reinitialize
port->sm_vars |= AD_PORT_BEGIN;
......@@ -2295,10 +2295,8 @@ void bond_3ad_handle_link_change(struct slave *slave, char link)
// if slave is null, the whole port is not initialized
if (!port->slave) {
#ifdef BONDING_DEBUG
printk(KERN_WARNING "bonding: Warning: link status changed for uninitialized port on %s\n",
printk(KERN_WARNING DRV_NAME ": Warning: link status changed for uninitialized port on %s\n",
slave->dev->name);
#endif
return;
}
......@@ -2356,41 +2354,27 @@ int bond_3ad_get_active_agg_info(struct bonding *bond, struct ad_info *ad_info)
int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
{
slave_t *slave, *start_at;
struct bonding *bond = (struct bonding *) dev->priv;
struct slave *slave, *start_at;
struct bonding *bond = dev->priv;
struct ethhdr *data = (struct ethhdr *)skb->data;
int slave_agg_no;
int slaves_in_agg;
int agg_id;
int i;
struct ad_info ad_info;
if (!IS_UP(dev)) { /* bond down */
dev_kfree_skb(skb);
return 0;
}
if (bond == NULL) {
printk(KERN_CRIT "bonding: Error: bond is NULL on device %s\n", dev->name);
dev_kfree_skb(skb);
return 0;
}
/* make sure that the slaves list will
* not change during tx
*/
read_lock(&bond->lock);
slave = bond->prev;
/* check if bond is empty */
if ((slave == (struct slave *) bond) || (bond->slave_cnt == 0)) {
printk(KERN_DEBUG "ERROR: bond is empty\n");
dev_kfree_skb(skb);
read_unlock(&bond->lock);
return 0;
if (!BOND_IS_OK(bond)) {
goto free_out;
}
if (bond_3ad_get_active_agg_info(bond, &ad_info)) {
printk(KERN_DEBUG "ERROR: bond_3ad_get_active_agg_info failed\n");
dev_kfree_skb(skb);
read_unlock(&bond->lock);
return 0;
goto free_out;
}
slaves_in_agg = ad_info.ports;
......@@ -2399,21 +2383,12 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
if (slaves_in_agg == 0) {
/*the aggregator is empty*/
printk(KERN_DEBUG "ERROR: active aggregator is empty\n");
dev_kfree_skb(skb);
read_unlock(&bond->lock);
return 0;
goto free_out;
}
/* we're at the root, get the first slave */
if ((slave == NULL) || (slave->dev == NULL)) {
/* no suitable interface, frame not sent */
dev_kfree_skb(skb);
read_unlock(&bond->lock);
return 0;
}
slave_agg_no = (data->h_dest[5]^bond->dev->dev_addr[5]) % slaves_in_agg;
slave_agg_no = (data->h_dest[5]^slave->dev->dev_addr[5]) % slaves_in_agg;
while (slave != (slave_t *)bond) {
bond_for_each_slave(bond, slave, i) {
struct aggregator *agg = SLAVE_AD_INFO(slave).port.aggregator;
if (agg && (agg->aggregator_identifier == agg_id)) {
......@@ -2422,37 +2397,18 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
break;
}
}
slave = slave->prev;
if (slave == NULL) {
printk(KERN_ERR "bonding: Error: slave is NULL\n");
dev_kfree_skb(skb);
read_unlock(&bond->lock);
return 0;
}
}
if (slave == (slave_t *)bond) {
printk(KERN_ERR "bonding: Error: Couldn't find a slave to tx on for aggregator ID %d\n", agg_id);
dev_kfree_skb(skb);
read_unlock(&bond->lock);
return 0;
if (slave_agg_no >= 0) {
printk(KERN_ERR DRV_NAME ": Error: Couldn't find a slave to tx on for aggregator ID %d\n", agg_id);
goto free_out;
}
start_at = slave;
do {
bond_for_each_slave_from(bond, slave, i, start_at) {
int slave_agg_id = 0;
struct aggregator *agg;
if (slave == NULL) {
printk(KERN_ERR "bonding: Error: slave is NULL\n");
dev_kfree_skb(skb);
read_unlock(&bond->lock);
return 0;
}
agg = SLAVE_AD_INFO(slave).port.aggregator;
struct aggregator *agg = SLAVE_AD_INFO(slave).port.aggregator;
if (agg) {
slave_agg_id = agg->aggregator_identifier;
......@@ -2463,20 +2419,24 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
skb->dev = slave->dev;
skb->priority = 1;
dev_queue_xmit(skb);
goto out;
}
}
out:
read_unlock(&bond->lock);
return 0;
}
} while ((slave = slave->next) != start_at);
free_out:
/* no suitable interface, frame not sent */
dev_kfree_skb(skb);
read_unlock(&bond->lock);
return 0;
goto out;
}
int bond_3ad_lacpdu_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type* ptype)
{
struct bonding *bond = (struct bonding *)dev->priv;
struct bonding *bond = dev->priv;
struct slave *slave = NULL;
int ret = NET_RX_DROP;
......
drivers/net/bonding/bond_3ad.h
View file @ 61b6a04a
......@@ -28,6 +28,9 @@
* 2003/05/01 - Shmulik Hen <shmulik.hen at intel dot com>
* - Renamed bond_3ad_link_status_changed() to
* bond_3ad_handle_link_change() for compatibility with TLB.
*
* 2003/09/24 - Shmulik Hen <shmulik.hen at intel dot com>
* - Code cleanup and style changes
*/
#ifndef __BOND_3AD_H__
......
drivers/net/bonding/bond_alb.c
View file @ 61b6a04a
......@@ -28,8 +28,13 @@
* 2003/08/06 - Amir Noam <amir.noam at intel dot com>
* - Add support for setting bond's MAC address with special
* handling required for ALB/TLB.
*
* 2003/09/24 - Shmulik Hen <shmulik.hen at intel dot com>
* - Code cleanup and style changes
*/
//#define BONDING_DEBUG 1
#include <linux/skbuff.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
......@@ -50,11 +55,11 @@
#define ALB_TIMER_TICKS_PER_SEC 10 /* should be a divisor of HZ */
#define BOND_TLB_REBALANCE_INTERVAL 10 /* in seconds, periodic re-balancing
* used for division - never set
#define BOND_TLB_REBALANCE_INTERVAL 10 /* In seconds, periodic re-balancing.
* Used for division - never set
* to zero !!!
*/
#define BOND_ALB_LP_INTERVAL 1 /* in seconds periodic send of
#define BOND_ALB_LP_INTERVAL 1 /* In seconds, periodic send of
* learning packets to the switch
*/
......@@ -66,7 +71,7 @@
#define TLB_HASH_TABLE_SIZE 256 /* The size of the clients hash table.
* Note that this value MUST NOT be smaller
* because the key hash table BYTE wide !
* because the key hash table is BYTE wide !
*/
......@@ -86,12 +91,15 @@
*/
#define RLB_PROMISC_TIMEOUT 10*ALB_TIMER_TICKS_PER_SEC
static const u8 mac_bcast[ETH_ALEN] = {0xff,0xff,0xff,0xff,0xff,0xff};
static const int alb_delta_in_ticks = HZ / ALB_TIMER_TICKS_PER_SEC;
#pragma pack(1)
struct learning_pkt {
u8 mac_dst[ETH_ALEN];
u8 mac_src[ETH_ALEN];
u16 type;
u8 padding[ETH_ZLEN - (2*ETH_ALEN + 2)];
u8 padding[ETH_ZLEN - ETH_HLEN];
};
struct arp_pkt {
......@@ -110,13 +118,12 @@ struct arp_pkt {
/* Forward declaration */
static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[]);
static inline u8
_simple_hash(u8 *hash_start, int hash_size)
static inline u8 _simple_hash(u8 *hash_start, int hash_size)
{
int i;
u8 hash = 0;
for (i=0; i<hash_size; i++) {
for (i = 0; i < hash_size; i++) {
hash ^= hash_start[i];
}
......@@ -125,193 +132,151 @@ _simple_hash(u8 *hash_start, int hash_size)
/*********************** tlb specific functions ***************************/
static inline void
_lock_tx_hashtbl(struct bonding *bond)
static inline void _lock_tx_hashtbl(struct bonding *bond)
{
spin_lock(&(BOND_ALB_INFO(bond).tx_hashtbl_lock));
}
static inline void
_unlock_tx_hashtbl(struct bonding *bond)
static inline void _unlock_tx_hashtbl(struct bonding *bond)
{
spin_unlock(&(BOND_ALB_INFO(bond).tx_hashtbl_lock));
}
/* Caller must hold tx_hashtbl lock */
static inline void
tlb_init_table_entry(struct bonding *bond, u8 index, u8 save_load)
static inline void tlb_init_table_entry(struct tlb_client_info *entry, int save_load)
{
struct tlb_client_info *entry;
if (BOND_ALB_INFO(bond).tx_hashtbl == NULL) {
return;
}
entry = &(BOND_ALB_INFO(bond).tx_hashtbl[index]);
/* at end of cycle, save the load that was transmitted to the client
* during the cycle, and set the tx_bytes counter to 0 for counting
* the load during the next cycle
*/
if (save_load) {
entry->load_history = 1 + entry->tx_bytes /
BOND_TLB_REBALANCE_INTERVAL;
entry->tx_bytes = 0;
}
entry->tx_slave = NULL;
entry->next = TLB_NULL_INDEX;
entry->prev = TLB_NULL_INDEX;
}
static inline void
tlb_init_slave(struct slave *slave)
static inline void tlb_init_slave(struct slave *slave)
{
struct tlb_slave_info *slave_info = &(SLAVE_TLB_INFO(slave));
slave_info->load = 0;
slave_info->head = TLB_NULL_INDEX;
SLAVE_TLB_INFO(slave).load = 0;
SLAVE_TLB_INFO(slave).head = TLB_NULL_INDEX;
}
/* Caller must hold bond lock for read */
static inline void
tlb_clear_slave(struct bonding *bond, struct slave *slave, u8 save_load)
static void tlb_clear_slave(struct bonding *bond, struct slave *slave, int save_load)
{
struct tlb_client_info *tx_hash_table = NULL;
u32 index, next_index;
struct tlb_client_info *tx_hash_table;
u32 index;
/* clear slave from tx_hashtbl */
_lock_tx_hashtbl(bond);
/* clear slave from tx_hashtbl */
tx_hash_table = BOND_ALB_INFO(bond).tx_hashtbl;
if (tx_hash_table) {
index = SLAVE_TLB_INFO(slave).head;
while (index != TLB_NULL_INDEX) {
next_index = tx_hash_table[index].next;
tlb_init_table_entry(bond, index, save_load);
u32 next_index = tx_hash_table[index].next;
tlb_init_table_entry(&tx_hash_table[index], save_load);
index = next_index;
}
}
_unlock_tx_hashtbl(bond);
tlb_init_slave(slave);
}
/* Must be called before starting the monitor timer */
static int
tlb_initialize(struct bonding *bond)
static int tlb_initialize(struct bonding *bond)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
int size = TLB_HASH_TABLE_SIZE * sizeof(struct tlb_client_info);
int i;
size_t size;
#if(TLB_HASH_TABLE_SIZE != 256)
/* Key to the hash table is byte wide. Check the size! */
#error Hash Table size is wrong.
#endif
spin_lock_init(&(bond_info->tx_hashtbl_lock));
_lock_tx_hashtbl(bond);
if (bond_info->tx_hashtbl != NULL) {
printk (KERN_ERR "%s: TLB hash table is not NULL\n",
bond->device->name);
_unlock_tx_hashtbl(bond);
return -1;
}
size = TLB_HASH_TABLE_SIZE * sizeof(struct tlb_client_info);
bond_info->tx_hashtbl = kmalloc(size, GFP_KERNEL);
if (bond_info->tx_hashtbl == NULL) {
printk (KERN_ERR "%s: Failed to allocate TLB hash table\n",
bond->device->name);
if (!bond_info->tx_hashtbl) {
printk(KERN_ERR DRV_NAME
": Error: %s: Failed to allocate TLB hash table\n",
bond->dev->name);
_unlock_tx_hashtbl(bond);
return -1;
}
memset(bond_info->tx_hashtbl, 0, size);
for (i=0; i<TLB_HASH_TABLE_SIZE; i++) {
tlb_init_table_entry(bond, i, 1);
for (i = 0; i < TLB_HASH_TABLE_SIZE; i++) {
tlb_init_table_entry(&bond_info->tx_hashtbl[i], 1);
}
_unlock_tx_hashtbl(bond);
return 0;
}
/* Must be called only after all slaves have been released */
static void
tlb_deinitialize(struct bonding *bond)
static void tlb_deinitialize(struct bonding *bond)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
_lock_tx_hashtbl(bond);
if (bond_info->tx_hashtbl == NULL) {
_unlock_tx_hashtbl(bond);
return;
}
kfree(bond_info->tx_hashtbl);
bond_info->tx_hashtbl = NULL;
_unlock_tx_hashtbl(bond);
}
/* Caller must hold bond lock for read */
static struct slave*
tlb_get_least_loaded_slave(struct bonding *bond)
static struct slave *tlb_get_least_loaded_slave(struct bonding *bond)
{
struct slave *slave;
struct slave *least_loaded;
s64 curr_gap, max_gap;
struct slave *slave, *least_loaded;
s64 max_gap;
int i, found = 0;
/* Find the first enabled slave */
slave = bond_get_first_slave(bond);
while (slave) {
bond_for_each_slave(bond, slave, i) {
if (SLAVE_IS_OK(slave)) {
found = 1;
break;
}
slave = bond_get_next_slave(bond, slave);
}
if (!slave) {
if (!found) {
return NULL;
}
least_loaded = slave;
max_gap = (s64)(slave->speed * 1000000) -
(s64)(SLAVE_TLB_INFO(slave).load * 8);
max_gap = (s64)(slave->speed << 20) - /* Convert to Megabit per sec */
(s64)(SLAVE_TLB_INFO(slave).load << 3); /* Bytes to bits */
/* Find the slave with the largest gap */
slave = bond_get_next_slave(bond, slave);
while (slave) {
bond_for_each_slave_from(bond, slave, i, least_loaded) {
if (SLAVE_IS_OK(slave)) {
curr_gap = (s64)(slave->speed * 1000000) -
(s64)(SLAVE_TLB_INFO(slave).load * 8);
if (max_gap < curr_gap) {
s64 gap = (s64)(slave->speed << 20) -
(s64)(SLAVE_TLB_INFO(slave).load << 3);
if (max_gap < gap) {
least_loaded = slave;
max_gap = curr_gap;
max_gap = gap;
}
}
slave = bond_get_next_slave(bond, slave);
}
return least_loaded;
}
/* Caller must hold bond lock for read */
struct slave*
tlb_choose_channel(struct bonding *bond, u32 hash_index, u32 skb_len)
struct slave *tlb_choose_channel(struct bonding *bond, u32 hash_index, u32 skb_len)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
struct tlb_client_info *hash_table = NULL;
struct slave *assigned_slave = NULL;
struct tlb_client_info *hash_table;
struct slave *assigned_slave;
_lock_tx_hashtbl(bond);
hash_table = bond_info->tx_hashtbl;
if (hash_table == NULL) {
printk (KERN_ERR "%s: TLB hash table is NULL\n",
bond->device->name);
_unlock_tx_hashtbl(bond);
return NULL;
}
assigned_slave = hash_table[hash_index].tx_slave;
if (!assigned_slave) {
assigned_slave = tlb_get_least_loaded_slave(bond);
......@@ -345,14 +310,12 @@ tlb_choose_channel(struct bonding *bond, u32 hash_index, u32 skb_len)
}
/*********************** rlb specific functions ***************************/
static inline void
_lock_rx_hashtbl(struct bonding *bond)
static inline void _lock_rx_hashtbl(struct bonding *bond)
{
spin_lock(&(BOND_ALB_INFO(bond).rx_hashtbl_lock));
}
static inline void
_unlock_rx_hashtbl(struct bonding *bond)
static inline void _unlock_rx_hashtbl(struct bonding *bond)
{
spin_unlock(&(BOND_ALB_INFO(bond).rx_hashtbl_lock));
}
......@@ -360,26 +323,20 @@ _unlock_rx_hashtbl(struct bonding *bond)
/* when an ARP REPLY is received from a client update its info
* in the rx_hashtbl
*/
static void
rlb_update_entry_from_arp(struct bonding *bond, struct arp_pkt *arp)
static void rlb_update_entry_from_arp(struct bonding *bond, struct arp_pkt *arp)
{
u32 hash_index;
struct rlb_client_info *client_info = NULL;
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
struct rlb_client_info *client_info;
u32 hash_index;
_lock_rx_hashtbl(bond);
if (bond_info->rx_hashtbl == NULL) {
_unlock_rx_hashtbl(bond);
return;
}
hash_index = _simple_hash((u8*)&(arp->ip_src), 4);
hash_index = _simple_hash((u8*)&(arp->ip_src), sizeof(arp->ip_src));
client_info = &(bond_info->rx_hashtbl[hash_index]);
if ((client_info->assigned) &&
(client_info->ip_src == arp->ip_dst) &&
(client_info->ip_dst == arp->ip_src)) {
/* update the clients MAC address */
memcpy(client_info->mac_dst, arp->mac_src, ETH_ALEN);
client_info->ntt = 1;
......@@ -389,66 +346,60 @@ rlb_update_entry_from_arp(struct bonding *bond, struct arp_pkt *arp)
_unlock_rx_hashtbl(bond);
}
static int
rlb_arp_recv(struct sk_buff *skb,
struct net_device *dev,
struct packet_type* ptype)
static int rlb_arp_recv(struct sk_buff *skb, struct net_device *bond_dev, struct packet_type *ptype)
{
struct bonding *bond = (struct bonding *)dev->priv;
int ret = NET_RX_DROP;
struct bonding *bond = bond_dev->priv;
struct arp_pkt *arp = (struct arp_pkt *)skb->data;
int res = NET_RX_DROP;
if (!(dev->flags & IFF_MASTER)) {
if (!(bond_dev->flags & IFF_MASTER)) {
goto out;
}
if (!arp) {
printk(KERN_ERR "Packet has no ARP data\n");
dprintk("Packet has no ARP data\n");
goto out;
}
if (skb->len < sizeof(struct arp_pkt)) {
printk(KERN_ERR "Packet is too small to be an ARP\n");
dprintk("Packet is too small to be an ARP\n");
goto out;
}
if (arp->op_code == htons(ARPOP_REPLY)) {
/* update rx hash table for this ARP */
rlb_update_entry_from_arp(bond, arp);
BOND_PRINT_DBG(("Server received an ARP Reply from client"));
dprintk("Server received an ARP Reply from client\n");
}
ret = NET_RX_SUCCESS;
res = NET_RX_SUCCESS;
out:
dev_kfree_skb(skb);
return ret;
return res;
}
/* Caller must hold bond lock for read */
static struct slave*
rlb_next_rx_slave(struct bonding *bond)
static struct slave *rlb_next_rx_slave(struct bonding *bond)
{
struct slave *rx_slave = NULL, *slave = NULL;
unsigned int i = 0;
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
struct slave *rx_slave, *slave, *start_at;
int i = 0;
slave = bond_info->next_rx_slave;
if (slave == NULL) {
slave = bond->next;
if (bond_info->next_rx_slave) {
start_at = bond_info->next_rx_slave;
} else {
start_at = bond->first_slave;
}
/* this loop uses the circular linked list property of the
* slave's list to go through all slaves
*/
for (i = 0; i < bond->slave_cnt; i++, slave = slave->next) {
rx_slave = NULL;
bond_for_each_slave_from(bond, slave, i, start_at) {
if (SLAVE_IS_OK(slave)) {
if (!rx_slave) {
rx_slave = slave;
}
else if (slave->speed > rx_slave->speed) {
} else if (slave->speed > rx_slave->speed) {
rx_slave = slave;
}
}
......@@ -464,48 +415,41 @@ rlb_next_rx_slave(struct bonding *bond)
/* teach the switch the mac of a disabled slave
* on the primary for fault tolerance
*
* Caller must hold bond->ptrlock for write or bond lock for write
* Caller must hold bond->curr_slave_lock for write or bond lock for write
*/
static void
rlb_teach_disabled_mac_on_primary(struct bonding *bond, u8 addr[])
static void rlb_teach_disabled_mac_on_primary(struct bonding *bond, u8 addr[])
{
if (!bond->current_slave) {
if (!bond->curr_active_slave) {
return;
}
if (!bond->alb_info.primary_is_promisc) {
bond->alb_info.primary_is_promisc = 1;
dev_set_promiscuity(bond->current_slave->dev, 1);
dev_set_promiscuity(bond->curr_active_slave->dev, 1);
}
bond->alb_info.rlb_promisc_timeout_counter = 0;
alb_send_learning_packets(bond->current_slave, addr);
alb_send_learning_packets(bond->curr_active_slave, addr);
}
/* slave being removed should not be active at this point
*
* Caller must hold bond lock for read
*/
static void
rlb_clear_slave(struct bonding *bond, struct slave *slave)
static void rlb_clear_slave(struct bonding *bond, struct slave *slave)
{
struct rlb_client_info *rx_hash_table = NULL;
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
u8 mac_bcast[ETH_ALEN] = {0xff,0xff,0xff,0xff,0xff,0xff};
struct rlb_client_info *rx_hash_table;
u32 index, next_index;
/* clear slave from rx_hashtbl */
_lock_rx_hashtbl(bond);
rx_hash_table = bond_info->rx_hashtbl;
if (rx_hash_table == NULL) {
_unlock_rx_hashtbl(bond);
return;
}
rx_hash_table = bond_info->rx_hashtbl;
index = bond_info->rx_hashtbl_head;
for (; index != RLB_NULL_INDEX; index = next_index) {
next_index = rx_hash_table[index].next;
if (rx_hash_table[index].slave == slave) {
struct slave *assigned_slave = rlb_next_rx_slave(bond);
......@@ -533,23 +477,24 @@ rlb_clear_slave(struct bonding *bond, struct slave *slave)
_unlock_rx_hashtbl(bond);
write_lock(&bond->ptrlock);
if (slave != bond->current_slave) {
write_lock(&bond->curr_slave_lock);
if (slave != bond->curr_active_slave) {
rlb_teach_disabled_mac_on_primary(bond, slave->dev->dev_addr);
}
write_unlock(&bond->ptrlock);
write_unlock(&bond->curr_slave_lock);
}
static void
rlb_update_client(struct rlb_client_info *client_info)
static void rlb_update_client(struct rlb_client_info *client_info)
{
int i = 0;
int i;
if (client_info->slave == NULL) {
if (!client_info->slave) {
return;
}
for (i=0; i<RLB_ARP_BURST_SIZE; i++) {
for (i = 0; i < RLB_ARP_BURST_SIZE; i++) {
arp_send(ARPOP_REPLY, ETH_P_ARP,
client_info->ip_dst,
client_info->slave->dev,
......@@ -561,20 +506,14 @@ rlb_update_client(struct rlb_client_info *client_info)
}
/* sends ARP REPLIES that update the clients that need updating */
static void
rlb_update_rx_clients(struct bonding *bond)
static void rlb_update_rx_clients(struct bonding *bond)
{
u32 hash_index;
struct rlb_client_info *client_info = NULL;
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
struct rlb_client_info *client_info;
u32 hash_index;
_lock_rx_hashtbl(bond);
if (bond_info->rx_hashtbl == NULL) {
_unlock_rx_hashtbl(bond);
return;
}
hash_index = bond_info->rx_hashtbl_head;
for (; hash_index != RLB_NULL_INDEX; hash_index = client_info->next) {
client_info = &(bond_info->rx_hashtbl[hash_index]);
......@@ -595,22 +534,15 @@ rlb_update_rx_clients(struct bonding *bond)
}
/* The slave was assigned a new mac address - update the clients */
static void
rlb_req_update_slave_clients(struct bonding *bond, struct slave *slave)
static void rlb_req_update_slave_clients(struct bonding *bond, struct slave *slave)
{
u32 hash_index;
u8 ntt = 0;
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
u8 mac_bcast[ETH_ALEN] = {0xff,0xff,0xff,0xff,0xff,0xff};
struct rlb_client_info* client_info = NULL;
struct rlb_client_info *client_info;
int ntt = 0;
u32 hash_index;
_lock_rx_hashtbl(bond);
if (bond_info->rx_hashtbl == NULL) {
_unlock_rx_hashtbl(bond);
return;
}
hash_index = bond_info->rx_hashtbl_head;
for (; hash_index != RLB_NULL_INDEX; hash_index = client_info->next) {
client_info = &(bond_info->rx_hashtbl[hash_index]);
......@@ -633,37 +565,31 @@ rlb_req_update_slave_clients(struct bonding *bond, struct slave *slave)
}
/* mark all clients using src_ip to be updated */
static void
rlb_req_update_subnet_clients(struct bonding *bond, u32 src_ip)
static void rlb_req_update_subnet_clients(struct bonding *bond, u32 src_ip)
{
u32 hash_index;
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
u8 mac_bcast[ETH_ALEN] = {0xff,0xff,0xff,0xff,0xff,0xff};
struct rlb_client_info *client_info = NULL;
struct rlb_client_info *client_info;
u32 hash_index;
_lock_rx_hashtbl(bond);
if (bond_info->rx_hashtbl == NULL) {
_unlock_rx_hashtbl(bond);
return;
}
hash_index = bond_info->rx_hashtbl_head;
for (; hash_index != RLB_NULL_INDEX; hash_index = client_info->next) {
client_info = &(bond_info->rx_hashtbl[hash_index]);
if (!client_info->slave) {
printk(KERN_ERR "Bonding: Error: found a client with no"
" channel in the client's hash table\n");
printk(KERN_ERR DRV_NAME
": Error: found a client with no channel in "
"the client's hash table\n");
continue;
}
/*update all clients using this src_ip, that are not assigned
* to the team's address (current_slave) and have a known
* to the team's address (curr_active_slave) and have a known
* unicast mac address.
*/
if ((client_info->ip_src == src_ip) &&
memcmp(client_info->slave->dev->dev_addr,
bond->device->dev_addr, ETH_ALEN) &&
bond->dev->dev_addr, ETH_ALEN) &&
memcmp(client_info->mac_dst, mac_bcast, ETH_ALEN)) {
client_info->ntt = 1;
bond_info->rx_ntt = 1;
......@@ -674,30 +600,22 @@ rlb_req_update_subnet_clients(struct bonding *bond, u32 src_ip)
}
/* Caller must hold both bond and ptr locks for read */
struct slave*
rlb_choose_channel(struct bonding *bond, struct arp_pkt *arp)
struct slave *rlb_choose_channel(struct bonding *bond, struct arp_pkt *arp)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
struct rlb_client_info *client_info = NULL;
struct slave *assigned_slave;
struct rlb_client_info *client_info;
u32 hash_index = 0;
struct slave *assigned_slave = NULL;
u8 mac_bcast[ETH_ALEN] = {0xff,0xff,0xff,0xff,0xff,0xff};
_lock_rx_hashtbl(bond);
if (bond_info->rx_hashtbl == NULL) {
_unlock_rx_hashtbl(bond);
return NULL;
}
hash_index = _simple_hash((u8 *)&arp->ip_dst, 4);
hash_index = _simple_hash((u8 *)&arp->ip_dst, sizeof(arp->ip_src));
client_info = &(bond_info->rx_hashtbl[hash_index]);
if (client_info->assigned == 1) {
if (client_info->assigned) {
if ((client_info->ip_src == arp->ip_src) &&
(client_info->ip_dst == arp->ip_dst)) {
/* the entry is already assigned to this client */
if (memcmp(arp->mac_dst, mac_bcast, ETH_ALEN)) {
/* update mac address from arp */
memcpy(client_info->mac_dst, arp->mac_dst, ETH_ALEN);
......@@ -710,12 +628,12 @@ rlb_choose_channel(struct bonding *bond, struct arp_pkt *arp)
}
} else {
/* the entry is already assigned to some other client,
* move the old client to primary (current_slave) so
* move the old client to primary (curr_active_slave) so
* that the new client can be assigned to this entry.
*/
if (bond->current_slave &&
client_info->slave != bond->current_slave) {
client_info->slave = bond->current_slave;
if (bond->curr_active_slave &&
client_info->slave != bond->curr_active_slave) {
client_info->slave = bond->curr_active_slave;
rlb_update_client(client_info);
}
}
......@@ -736,8 +654,7 @@ rlb_choose_channel(struct bonding *bond, struct arp_pkt *arp)
if (memcmp(client_info->mac_dst, mac_bcast, ETH_ALEN)) {
client_info->ntt = 1;
bond->alb_info.rx_ntt = 1;
}
else {
} else {
client_info->ntt = 0;
}
......@@ -760,10 +677,9 @@ rlb_choose_channel(struct bonding *bond, struct arp_pkt *arp)
/* chooses (and returns) transmit channel for arp reply
* does not choose channel for other arp types since they are
* sent on the current_slave
* sent on the curr_active_slave
*/
static struct slave*
rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
{
struct arp_pkt *arp = (struct arp_pkt *)skb->nh.raw;
struct slave *tx_slave = NULL;
......@@ -776,9 +692,8 @@ rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
if (tx_slave) {
memcpy(arp->mac_src,tx_slave->dev->dev_addr, ETH_ALEN);
}
BOND_PRINT_DBG(("Server sent ARP Reply packet"));
dprintk("Server sent ARP Reply packet\n");
} else if (arp->op_code == __constant_htons(ARPOP_REQUEST)) {
/* Create an entry in the rx_hashtbl for this client as a
* place holder.
* When the arp reply is received the entry will be updated
......@@ -797,34 +712,29 @@ rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
* updated with their assigned mac.
*/
rlb_req_update_subnet_clients(bond, arp->ip_src);
BOND_PRINT_DBG(("Server sent ARP Request packet"));
dprintk("Server sent ARP Request packet\n");
}
return tx_slave;
}
/* Caller must hold bond lock for read */
static void
rlb_rebalance(struct bonding *bond)
static void rlb_rebalance(struct bonding *bond)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
struct slave *assigned_slave = NULL;
struct slave *assigned_slave;
struct rlb_client_info *client_info;
int ntt;
u32 hash_index;
struct rlb_client_info *client_info = NULL;
u8 ntt = 0;
_lock_rx_hashtbl(bond);
if (bond_info->rx_hashtbl == NULL) {
_unlock_rx_hashtbl(bond);
return;
}
ntt = 0;
hash_index = bond_info->rx_hashtbl_head;
for (; hash_index != RLB_NULL_INDEX; hash_index = client_info->next) {
client_info = &(bond_info->rx_hashtbl[hash_index]);
assigned_slave = rlb_next_rx_slave(bond);
if (assigned_slave && (client_info->slave != assigned_slave)){
if (assigned_slave && (client_info->slave != assigned_slave)) {
client_info->slave = assigned_slave;
client_info->ntt = 1;
ntt = 1;
......@@ -839,96 +749,83 @@ rlb_rebalance(struct bonding *bond)
}
/* Caller must hold rx_hashtbl lock */
static inline void
rlb_init_table_entry(struct rlb_client_info *entry)
static void rlb_init_table_entry(struct rlb_client_info *entry)
{
memset(entry, 0, sizeof(struct rlb_client_info));
entry->next = RLB_NULL_INDEX;
entry->prev = RLB_NULL_INDEX;
entry->assigned = 0;
entry->ntt = 0;
}
static int
rlb_initialize(struct bonding *bond)
static int rlb_initialize(struct bonding *bond)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
struct packet_type *pk_type = &(BOND_ALB_INFO(bond).rlb_pkt_type);
int size = RLB_HASH_TABLE_SIZE * sizeof(struct rlb_client_info);
int i;
size_t size;
spin_lock_init(&(bond_info->rx_hashtbl_lock));
_lock_rx_hashtbl(bond);
if (bond_info->rx_hashtbl != NULL) {
printk (KERN_ERR "%s: RLB hash table is not NULL\n",
bond->device->name);
_unlock_rx_hashtbl(bond);
return -1;
}
size = RLB_HASH_TABLE_SIZE * sizeof(struct rlb_client_info);
bond_info->rx_hashtbl = kmalloc(size, GFP_KERNEL);
if (bond_info->rx_hashtbl == NULL) {
printk (KERN_ERR "%s: Failed to allocate"
" RLB hash table\n", bond->device->name);
if (!bond_info->rx_hashtbl) {
printk(KERN_ERR DRV_NAME
": Error: %s: Failed to allocate RLB hash table\n",
bond->dev->name);
_unlock_rx_hashtbl(bond);
return -1;
}
bond_info->rx_hashtbl_head = RLB_NULL_INDEX;
for (i=0; i<RLB_HASH_TABLE_SIZE; i++) {
for (i = 0; i < RLB_HASH_TABLE_SIZE; i++) {
rlb_init_table_entry(bond_info->rx_hashtbl + i);
}
_unlock_rx_hashtbl(bond);
/* register to receive ARPs */
_unlock_rx_hashtbl(bond);
/*initialize packet type*/
pk_type->type = __constant_htons(ETH_P_ARP);
pk_type->dev = bond->device;
pk_type->dev = bond->dev;
pk_type->func = rlb_arp_recv;
/* register to receive ARPs */
dev_add_pack(pk_type);
return 0;
}
static void
rlb_deinitialize(struct bonding *bond)
static void rlb_deinitialize(struct bonding *bond)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
dev_remove_pack(&(bond_info->rlb_pkt_type));
_lock_rx_hashtbl(bond);
if (bond_info->rx_hashtbl == NULL) {
_unlock_rx_hashtbl(bond);
return;
}
kfree(bond_info->rx_hashtbl);
bond_info->rx_hashtbl = NULL;
_unlock_rx_hashtbl(bond);
}
/*********************** tlb/rlb shared functions *********************/
static void
alb_send_learning_packets(struct slave *slave, u8 mac_addr[])
static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[])
{
struct sk_buff *skb = NULL;
struct learning_pkt pkt;
char *data = NULL;
int size = sizeof(struct learning_pkt);
int i;
unsigned int size = sizeof(struct learning_pkt);
memset(&pkt, 0, size);
memcpy(pkt.mac_dst, mac_addr, ETH_ALEN);
memcpy(pkt.mac_src, mac_addr, ETH_ALEN);
pkt.type = __constant_htons(ETH_P_LOOP);
for (i=0; i < MAX_LP_RETRY; i++) {
skb = NULL;
for (i = 0; i < MAX_LP_RETRY; i++) {
struct sk_buff *skb;
char *data;
skb = dev_alloc_skb(size);
if (!skb) {
return;
......@@ -936,28 +833,26 @@ alb_send_learning_packets(struct slave *slave, u8 mac_addr[])
data = skb_put(skb, size);
memcpy(data, &pkt, size);
skb->mac.raw = data;
skb->nh.raw = data + ETH_HLEN;
skb->protocol = pkt.type;
skb->priority = TC_PRIO_CONTROL;
skb->dev = slave->dev;
dev_queue_xmit(skb);
}
}
/* hw is a boolean parameter that determines whether we should try and
* set the hw address of the device as well as the hw address of the
* net_device
*/
static int
alb_set_slave_mac_addr(struct slave *slave, u8 addr[], int hw)
static int alb_set_slave_mac_addr(struct slave *slave, u8 addr[], int hw)
{
struct net_device *dev = NULL;
struct net_device *dev = slave->dev;
struct sockaddr s_addr;
dev = slave->dev;
if (!hw) {
memcpy(dev->dev_addr, addr, dev->addr_len);
return 0;
......@@ -968,26 +863,23 @@ alb_set_slave_mac_addr(struct slave *slave, u8 addr[], int hw)
memcpy(s_addr.sa_data, addr, dev->addr_len);
s_addr.sa_family = dev->type;
if (dev->set_mac_address(dev, &s_addr)) {
printk(KERN_DEBUG "bonding: Error: alb_set_slave_mac_addr:"
" dev->set_mac_address of dev %s failed!"
" ALB mode requires that the base driver"
" support setting the hw address also when"
" the network device's interface is open\n",
printk(KERN_ERR DRV_NAME
": Error: dev->set_mac_address of dev %s failed! ALB "
"mode requires that the base driver support setting "
"the hw address also when the network device's "
"interface is open\n",
dev->name);
return -EOPNOTSUPP;
}
return 0;
}
/* Caller must hold bond lock for write or ptrlock for write*/
static void
alb_swap_mac_addr(struct bonding *bond,
struct slave *slave1,
struct slave *slave2)
/* Caller must hold bond lock for write or curr_slave_lock for write*/
static void alb_swap_mac_addr(struct bonding *bond, struct slave *slave1, struct slave *slave2)
{
u8 tmp_mac_addr[ETH_ALEN];
struct slave *disabled_slave = NULL;
u8 slaves_state_differ;
u8 tmp_mac_addr[ETH_ALEN];
int slaves_state_differ;
slaves_state_differ = (SLAVE_IS_OK(slave1) != SLAVE_IS_OK(slave2));
......@@ -1004,8 +896,7 @@ alb_swap_mac_addr(struct bonding *bond,
*/
rlb_req_update_slave_clients(bond, slave1);
}
}
else {
} else {
disabled_slave = slave1;
}
......@@ -1017,8 +908,7 @@ alb_swap_mac_addr(struct bonding *bond,
*/
rlb_req_update_slave_clients(bond, slave2);
}
}
else {
} else {
disabled_slave = slave2;
}
......@@ -1043,10 +933,8 @@ alb_swap_mac_addr(struct bonding *bond,
*
* Caller must hold bond lock
*/
static void
alb_change_hw_addr_on_detach(struct bonding *bond, struct slave *slave)
static void alb_change_hw_addr_on_detach(struct bonding *bond, struct slave *slave)
{
struct slave *tmp_slave;
int perm_curr_diff;
int perm_bond_diff;
......@@ -1054,20 +942,23 @@ alb_change_hw_addr_on_detach(struct bonding *bond, struct slave *slave)
slave->dev->dev_addr,
ETH_ALEN);
perm_bond_diff = memcmp(slave->perm_hwaddr,
bond->device->dev_addr,
bond->dev->dev_addr,
ETH_ALEN);
if (perm_curr_diff && perm_bond_diff) {
tmp_slave = bond_get_first_slave(bond);
while (tmp_slave) {
struct slave *tmp_slave;
int i, found = 0;
bond_for_each_slave(bond, tmp_slave, i) {
if (!memcmp(slave->perm_hwaddr,
tmp_slave->dev->dev_addr,
ETH_ALEN)) {
found = 1;
break;
}
tmp_slave = bond_get_next_slave(bond, tmp_slave);
}
if (tmp_slave) {
if (found) {
alb_swap_mac_addr(bond, slave, tmp_slave);
}
}
......@@ -1098,10 +989,10 @@ alb_change_hw_addr_on_detach(struct bonding *bond, struct slave *slave)
* caller must hold the bond lock for write since the mac addresses are compared
* and may be swapped.
*/
static int
alb_handle_addr_collision_on_attach(struct bonding *bond, struct slave *slave)
static int alb_handle_addr_collision_on_attach(struct bonding *bond, struct slave *slave)
{
struct slave *tmp_slave1, *tmp_slave2;
struct slave *tmp_slave1, *tmp_slave2, *free_mac_slave;
int i, j, found = 0;
if (bond->slave_cnt == 0) {
/* this is the first slave */
......@@ -1112,65 +1003,68 @@ alb_handle_addr_collision_on_attach(struct bonding *bond, struct slave *slave)
* check uniqueness of slave's mac address against the other
* slaves in the bond.
*/
if (memcmp(slave->perm_hwaddr, bond->device->dev_addr, ETH_ALEN)) {
tmp_slave1 = bond_get_first_slave(bond);
for (; tmp_slave1; tmp_slave1 = bond_get_next_slave(bond, tmp_slave1)) {
if (memcmp(slave->perm_hwaddr, bond->dev->dev_addr, ETH_ALEN)) {
bond_for_each_slave(bond, tmp_slave1, i) {
if (!memcmp(tmp_slave1->dev->dev_addr, slave->dev->dev_addr,
ETH_ALEN)) {
found = 1;
break;
}
}
if (tmp_slave1) {
if (found) {
/* a slave was found that is using the mac address
* of the new slave
*/
printk(KERN_ERR "bonding: Warning: the hw address "
"of slave %s is not unique - cannot enslave it!"
, slave->dev->name);
printk(KERN_ERR DRV_NAME
": Error: the hw address of slave %s is not "
"unique - cannot enslave it!",
slave->dev->name);
return -EINVAL;
}
return 0;
}
/* the slave's address is equal to the address of the bond
* search for a spare address in the bond for this slave.
/* The slave's address is equal to the address of the bond.
* Search for a spare address in the bond for this slave.
*/
tmp_slave1 = bond_get_first_slave(bond);
for (; tmp_slave1; tmp_slave1 = bond_get_next_slave(bond, tmp_slave1)) {
tmp_slave2 = bond_get_first_slave(bond);
for (; tmp_slave2; tmp_slave2 = bond_get_next_slave(bond, tmp_slave2)) {
free_mac_slave = NULL;
bond_for_each_slave(bond, tmp_slave1, i) {
found = 0;
bond_for_each_slave(bond, tmp_slave2, j) {
if (!memcmp(tmp_slave1->perm_hwaddr,
tmp_slave2->dev->dev_addr,
ETH_ALEN)) {
found = 1;
break;
}
}
if (!tmp_slave2) {
if (!found) {
/* no slave has tmp_slave1's perm addr
* as its curr addr
*/
free_mac_slave = tmp_slave1;
break;
}
}
if (tmp_slave1) {
alb_set_slave_mac_addr(slave, tmp_slave1->perm_hwaddr,
if (free_mac_slave) {
alb_set_slave_mac_addr(slave, free_mac_slave->perm_hwaddr,
bond->alb_info.rlb_enabled);
printk(KERN_WARNING "bonding: Warning: the hw address "
"of slave %s is in use by the bond; "
"giving it the hw address of %s\n",
slave->dev->name, tmp_slave1->dev->name);
printk(KERN_WARNING DRV_NAME
": Warning: the hw address of slave %s is in use by "
"the bond; giving it the hw address of %s\n",
slave->dev->name, free_mac_slave->dev->name);
} else {
printk(KERN_CRIT "bonding: Error: the hw address "
"of slave %s is in use by the bond; "
"couldn't find a slave with a free hw "
"address to give it (this should not have "
"happened)\n", slave->dev->name);
printk(KERN_ERR DRV_NAME
": Error: the hw address of slave %s is in use by the "
"bond; couldn't find a slave with a free hw address to "
"give it (this should not have happened)\n",
slave->dev->name);
return -EFAULT;
}
......@@ -1191,34 +1085,33 @@ alb_handle_addr_collision_on_attach(struct bonding *bond, struct slave *slave)
*
* Unwinding assumes bond's mac address has not yet changed.
*/
static inline int
alb_set_mac_address(struct bonding *bond, void *addr)
static int alb_set_mac_address(struct bonding *bond, void *addr)
{
struct sockaddr sa;
struct slave *slave;
struct slave *slave, *stop_at;
char tmp_addr[ETH_ALEN];
int error;
int res;
int i;
if (bond->alb_info.rlb_enabled) {
return 0;
}
slave = bond_get_first_slave(bond);
for (; slave; slave = bond_get_next_slave(bond, slave)) {
bond_for_each_slave(bond, slave, i) {
if (slave->dev->set_mac_address == NULL) {
error = -EOPNOTSUPP;
res = -EOPNOTSUPP;
goto unwind;
}
/* save net_device's current hw address */
memcpy(tmp_addr, slave->dev->dev_addr, ETH_ALEN);
error = slave->dev->set_mac_address(slave->dev, addr);
res = slave->dev->set_mac_address(slave->dev, addr);
/* restore net_device's hw address */
memcpy(slave->dev->dev_addr, tmp_addr, ETH_ALEN);
if (error) {
if (res) {
goto unwind;
}
}
......@@ -1226,22 +1119,23 @@ alb_set_mac_address(struct bonding *bond, void *addr)
return 0;
unwind:
memcpy(sa.sa_data, bond->device->dev_addr, bond->device->addr_len);
sa.sa_family = bond->device->type;
slave = bond_get_first_slave(bond);
for (; slave; slave = bond_get_next_slave(bond, slave)) {
memcpy(sa.sa_data, bond->dev->dev_addr, bond->dev->addr_len);
sa.sa_family = bond->dev->type;
/* unwind from head to the slave that failed */
stop_at = slave;
bond_for_each_slave_from_to(bond, slave, i, bond->first_slave, stop_at) {
memcpy(tmp_addr, slave->dev->dev_addr, ETH_ALEN);
slave->dev->set_mac_address(slave->dev, &sa);
memcpy(slave->dev->dev_addr, tmp_addr, ETH_ALEN);
}
return error;
return res;
}
/************************ exported alb funcions ************************/
int
bond_alb_initialize(struct bonding *bond, int rlb_enabled)
int bond_alb_initialize(struct bonding *bond, int rlb_enabled)
{
int res;
......@@ -1263,8 +1157,7 @@ bond_alb_initialize(struct bonding *bond, int rlb_enabled)
return 0;
}
void
bond_alb_deinitialize(struct bonding *bond)
void bond_alb_deinitialize(struct bonding *bond)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
......@@ -1275,49 +1168,38 @@ bond_alb_deinitialize(struct bonding *bond)
}
}
int
bond_alb_xmit(struct sk_buff *skb, struct net_device *dev)
int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
{
struct bonding *bond = (struct bonding *) dev->priv;
struct ethhdr *eth_data = (struct ethhdr *)skb->data;
struct bonding *bond = bond_dev->priv;
struct ethhdr *eth_data = (struct ethhdr *)skb->mac.raw = skb->data;
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
struct slave *tx_slave = NULL;
char do_tx_balance = 1;
static u32 ip_bcast = 0xffffffff;
int hash_size = 0;
int do_tx_balance = 1;
u32 hash_index = 0;
u8 *hash_start = NULL;
u8 mac_bcast[ETH_ALEN] = {0xff,0xff,0xff,0xff,0xff,0xff};
if (!IS_UP(dev)) { /* bond down */
dev_kfree_skb(skb);
return 0;
}
/* make sure that the current_slave and the slaves list do
/* make sure that the curr_active_slave and the slaves list do
* not change during tx
*/
read_lock(&bond->lock);
read_lock(&bond->curr_slave_lock);
if (bond->slave_cnt == 0) {
/* no suitable interface, frame not sent */
dev_kfree_skb(skb);
read_unlock(&bond->lock);
return 0;
if (!BOND_IS_OK(bond)) {
goto free_out;
}
read_lock(&bond->ptrlock);
switch (ntohs(skb->protocol)) {
case ETH_P_IP:
if ((memcmp(eth_data->h_dest, mac_bcast, ETH_ALEN) == 0) ||
(skb->nh.iph->daddr == 0xffffffff)) {
(skb->nh.iph->daddr == ip_bcast)) {
do_tx_balance = 0;
break;
}
hash_start = (char*)&(skb->nh.iph->daddr);
hash_size = 4;
hash_size = sizeof(skb->nh.iph->daddr);
break;
case ETH_P_IPV6:
if (memcmp(eth_data->h_dest, mac_bcast, ETH_ALEN) == 0) {
do_tx_balance = 0;
......@@ -1325,9 +1207,8 @@ bond_alb_xmit(struct sk_buff *skb, struct net_device *dev)
}
hash_start = (char*)&(skb->nh.ipv6h->daddr);
hash_size = 16;
hash_size = sizeof(skb->nh.ipv6h->daddr);
break;
case ETH_P_IPX:
if (ipx_hdr(skb)->ipx_checksum !=
__constant_htons(IPX_NO_CHECKSUM)) {
......@@ -1349,14 +1230,12 @@ bond_alb_xmit(struct sk_buff *skb, struct net_device *dev)
hash_start = (char*)eth_data->h_dest;
hash_size = ETH_ALEN;
break;
case ETH_P_ARP:
do_tx_balance = 0;
if (bond_info->rlb_enabled) {
tx_slave = rlb_arp_xmit(skb, bond);
}
break;
default:
do_tx_balance = 0;
break;
......@@ -1369,13 +1248,13 @@ bond_alb_xmit(struct sk_buff *skb, struct net_device *dev)
if (!tx_slave) {
/* unbalanced or unassigned, send through primary */
tx_slave = bond->current_slave;
tx_slave = bond->curr_active_slave;
bond_info->unbalanced_load += skb->len;
}
if (tx_slave && SLAVE_IS_OK(tx_slave)) {
skb->dev = tx_slave->dev;
if (tx_slave != bond->current_slave) {
if (tx_slave != bond->curr_active_slave) {
memcpy(eth_data->h_source,
tx_slave->dev->dev_addr,
ETH_ALEN);
......@@ -1386,26 +1265,35 @@ bond_alb_xmit(struct sk_buff *skb, struct net_device *dev)
if (tx_slave) {
tlb_clear_slave(bond, tx_slave, 0);
}
dev_kfree_skb(skb);
goto free_out;
}
read_unlock(&bond->ptrlock);
out:
read_unlock(&bond->curr_slave_lock);
read_unlock(&bond->lock);
return 0;
free_out:
dev_kfree_skb(skb);
goto out;
}
void
bond_alb_monitor(struct bonding *bond)
void bond_alb_monitor(struct bonding *bond)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
struct slave *slave = NULL;
struct slave *slave;
int i;
read_lock(&bond->lock);
if ((bond->slave_cnt == 0) || !(bond->device->flags & IFF_UP)) {
if (bond->kill_timers) {
goto out;
}
if (bond->slave_cnt == 0) {
bond_info->tx_rebalance_counter = 0;
bond_info->lp_counter = 0;
goto out;
goto re_arm;
}
bond_info->tx_rebalance_counter++;
......@@ -1413,51 +1301,53 @@ bond_alb_monitor(struct bonding *bond)
/* send learning packets */
if (bond_info->lp_counter >= BOND_ALB_LP_TICKS) {
/* change of current_slave involves swapping of mac addresses.
/* change of curr_active_slave involves swapping of mac addresses.
* in order to avoid this swapping from happening while
* sending the learning packets, the ptrlock must be held for
* sending the learning packets, the curr_slave_lock must be held for
* read.
*/
read_lock(&bond->ptrlock);
slave = bond_get_first_slave(bond);
while (slave) {
read_lock(&bond->curr_slave_lock);
bond_for_each_slave(bond, slave, i) {
alb_send_learning_packets(slave,slave->dev->dev_addr);
slave = bond_get_next_slave(bond, slave);
}
read_unlock(&bond->ptrlock);
read_unlock(&bond->curr_slave_lock);
bond_info->lp_counter = 0;
}
/* rebalance tx traffic */
if (bond_info->tx_rebalance_counter >= BOND_TLB_REBALANCE_TICKS) {
read_lock(&bond->ptrlock);
slave = bond_get_first_slave(bond);
while (slave) {
read_lock(&bond->curr_slave_lock);
bond_for_each_slave(bond, slave, i) {
tlb_clear_slave(bond, slave, 1);
if (slave == bond->current_slave) {
if (slave == bond->curr_active_slave) {
SLAVE_TLB_INFO(slave).load =
bond_info->unbalanced_load /
BOND_TLB_REBALANCE_INTERVAL;
bond_info->unbalanced_load = 0;
}
slave = bond_get_next_slave(bond, slave);
}
read_unlock(&bond->ptrlock);
read_unlock(&bond->curr_slave_lock);
bond_info->tx_rebalance_counter = 0;
}
/* handle rlb stuff */
if (bond_info->rlb_enabled) {
/* the following code changes the promiscuity of the
* the current_slave. It needs to be locked with a
* the curr_active_slave. It needs to be locked with a
* write lock to protect from other code that also
* sets the promiscuity.
*/
write_lock(&bond->ptrlock);
write_lock(&bond->curr_slave_lock);
if (bond_info->primary_is_promisc &&
(++bond_info->rlb_promisc_timeout_counter >=
RLB_PROMISC_TIMEOUT)) {
(++bond_info->rlb_promisc_timeout_counter >= RLB_PROMISC_TIMEOUT)) {
bond_info->rlb_promisc_timeout_counter = 0;
......@@ -1465,12 +1355,13 @@ bond_alb_monitor(struct bonding *bond)
* because a slave was disabled then
* it can now leave promiscuous mode.
*/
dev_set_promiscuity(bond->current_slave->dev, -1);
dev_set_promiscuity(bond->curr_active_slave->dev, -1);
bond_info->primary_is_promisc = 0;
}
write_unlock(&bond->ptrlock);
if (bond_info->rlb_rebalance == 1) {
write_unlock(&bond->curr_slave_lock);
if (bond_info->rlb_rebalance) {
bond_info->rlb_rebalance = 0;
rlb_rebalance(bond);
}
......@@ -1490,28 +1381,23 @@ bond_alb_monitor(struct bonding *bond)
}
}
re_arm:
mod_timer(&(bond_info->alb_timer), jiffies + alb_delta_in_ticks);
out:
read_unlock(&bond->lock);
if (bond->device->flags & IFF_UP) {
/* re-arm the timer */
mod_timer(&(bond_info->alb_timer),
jiffies + (HZ/ALB_TIMER_TICKS_PER_SEC));
}
}
/* assumption: called before the slave is attched to the bond
/* assumption: called before the slave is attached to the bond
* and not locked by the bond lock
*/
int
bond_alb_init_slave(struct bonding *bond, struct slave *slave)
int bond_alb_init_slave(struct bonding *bond, struct slave *slave)
{
int err = 0;
int res;
err = alb_set_slave_mac_addr(slave, slave->perm_hwaddr,
res = alb_set_slave_mac_addr(slave, slave->perm_hwaddr,
bond->alb_info.rlb_enabled);
if (err) {
return err;
if (res) {
return res;
}
/* caller must hold the bond lock for write since the mac addresses
......@@ -1519,12 +1405,12 @@ bond_alb_init_slave(struct bonding *bond, struct slave *slave)
*/
write_lock_bh(&bond->lock);
err = alb_handle_addr_collision_on_attach(bond, slave);
res = alb_handle_addr_collision_on_attach(bond, slave);
write_unlock_bh(&bond->lock);
if (err) {
return err;
if (res) {
return res;
}
tlb_init_slave(slave);
......@@ -1540,8 +1426,7 @@ bond_alb_init_slave(struct bonding *bond, struct slave *slave)
}
/* Caller must hold bond lock for write */
void
bond_alb_deinit_slave(struct bonding *bond, struct slave *slave)
void bond_alb_deinit_slave(struct bonding *bond, struct slave *slave)
{
if (bond->slave_cnt > 1) {
alb_change_hw_addr_on_detach(bond, slave);
......@@ -1556,9 +1441,7 @@ bond_alb_deinit_slave(struct bonding *bond, struct slave *slave)
}
/* Caller must hold bond lock for read */
void
bond_alb_handle_link_change(struct bonding *bond, struct slave *slave,
char link)
void bond_alb_handle_link_change(struct bonding *bond, struct slave *slave, char link)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
......@@ -1582,109 +1465,111 @@ bond_alb_handle_link_change(struct bonding *bond, struct slave *slave,
}
/**
* bond_alb_assign_current_slave - assign new current_slave
* bond_alb_handle_active_change - assign new curr_active_slave
* @bond: our bonding struct
* @new_slave: new slave to assign
*
* Set the bond->current_slave to @new_slave and handle
* Set the bond->curr_active_slave to @new_slave and handle
* mac address swapping and promiscuity changes as needed.
*
* Caller must hold bond ptrlock for write (or bond lock for write)
* Caller must hold bond curr_slave_lock for write (or bond lock for write)
*/
void
bond_alb_assign_current_slave(struct bonding *bond, struct slave *new_slave)
void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave)
{
struct slave *swap_slave = bond->current_slave;
struct slave *swap_slave;
int i;
if (bond->current_slave == new_slave) {
if (bond->curr_active_slave == new_slave) {
return;
}
if (bond->current_slave && bond->alb_info.primary_is_promisc) {
dev_set_promiscuity(bond->current_slave->dev, -1);
if (bond->curr_active_slave && bond->alb_info.primary_is_promisc) {
dev_set_promiscuity(bond->curr_active_slave->dev, -1);
bond->alb_info.primary_is_promisc = 0;
bond->alb_info.rlb_promisc_timeout_counter = 0;
}
bond->current_slave = new_slave;
swap_slave = bond->curr_active_slave;
bond->curr_active_slave = new_slave;
if (!new_slave || (bond->slave_cnt == 0)) {
return;
}
/* set the new current_slave to the bonds mac address
* i.e. swap mac addresses of old current_slave and new current_slave
/* set the new curr_active_slave to the bonds mac address
* i.e. swap mac addresses of old curr_active_slave and new curr_active_slave
*/
if (!swap_slave) {
struct slave *tmp_slave;
/* find slave that is holding the bond's mac address */
swap_slave = bond_get_first_slave(bond);
while (swap_slave) {
if (!memcmp(swap_slave->dev->dev_addr,
bond->device->dev_addr, ETH_ALEN)) {
bond_for_each_slave(bond, tmp_slave, i) {
if (!memcmp(tmp_slave->dev->dev_addr,
bond->dev->dev_addr, ETH_ALEN)) {
swap_slave = tmp_slave;
break;
}
swap_slave = bond_get_next_slave(bond, swap_slave);
}
}
/* current_slave must be set before calling alb_swap_mac_addr */
/* curr_active_slave must be set before calling alb_swap_mac_addr */
if (swap_slave) {
/* swap mac address */
alb_swap_mac_addr(bond, swap_slave, new_slave);
} else {
/* set the new_slave to the bond mac address */
alb_set_slave_mac_addr(new_slave, bond->device->dev_addr,
alb_set_slave_mac_addr(new_slave, bond->dev->dev_addr,
bond->alb_info.rlb_enabled);
/* fasten bond mac on new current slave */
alb_send_learning_packets(new_slave, bond->device->dev_addr);
alb_send_learning_packets(new_slave, bond->dev->dev_addr);
}
}
int
bond_alb_set_mac_address(struct net_device *dev, void *addr)
int bond_alb_set_mac_address(struct net_device *bond_dev, void *addr)
{
struct bonding *bond = dev->priv;
struct bonding *bond = bond_dev->priv;
struct sockaddr *sa = addr;
struct slave *swap_slave = NULL;
int error = 0;
struct slave *slave, *swap_slave;
int res;
int i;
if (!is_valid_ether_addr(sa->sa_data)) {
return -EADDRNOTAVAIL;
}
error = alb_set_mac_address(bond, addr);
if (error) {
return error;
res = alb_set_mac_address(bond, addr);
if (res) {
return res;
}
memcpy(dev->dev_addr, sa->sa_data, dev->addr_len);
memcpy(bond_dev->dev_addr, sa->sa_data, bond_dev->addr_len);
/* If there is no current_slave there is nothing else to do.
/* If there is no curr_active_slave there is nothing else to do.
* Otherwise we'll need to pass the new address to it and handle
* duplications.
*/
if (bond->current_slave == NULL) {
if (!bond->curr_active_slave) {
return 0;
}
swap_slave = bond_get_first_slave(bond);
while (swap_slave) {
if (!memcmp(swap_slave->dev->dev_addr, dev->dev_addr, ETH_ALEN)) {
swap_slave = NULL;
bond_for_each_slave(bond, slave, i) {
if (!memcmp(slave->dev->dev_addr, bond_dev->dev_addr, ETH_ALEN)) {
swap_slave = slave;
break;
}
swap_slave = bond_get_next_slave(bond, swap_slave);
}
if (swap_slave) {
alb_swap_mac_addr(bond, swap_slave, bond->current_slave);
alb_swap_mac_addr(bond, swap_slave, bond->curr_active_slave);
} else {
alb_set_slave_mac_addr(bond->current_slave, dev->dev_addr,
alb_set_slave_mac_addr(bond->curr_active_slave, bond_dev->dev_addr,
bond->alb_info.rlb_enabled);
alb_send_learning_packets(bond->current_slave, dev->dev_addr);
alb_send_learning_packets(bond->curr_active_slave, bond_dev->dev_addr);
if (bond->alb_info.rlb_enabled) {
/* inform clients mac address has changed */
rlb_req_update_slave_clients(bond, bond->current_slave);
rlb_req_update_slave_clients(bond, bond->curr_active_slave);
}
}
......
drivers/net/bonding/bond_alb.h
View file @ 61b6a04a
......@@ -24,6 +24,9 @@
* 2003/08/06 - Amir Noam <amir.noam at intel dot com>
* - Add support for setting bond's MAC address with special
* handling required for ALB/TLB.
*
* 2003/09/24 - Shmulik Hen <shmulik.hen at intel dot com>
* - Code cleanup and style changes
*/
#ifndef __BOND_ALB_H__
......@@ -126,10 +129,10 @@ void bond_alb_deinitialize(struct bonding *bond);
int bond_alb_init_slave(struct bonding *bond, struct slave *slave);
void bond_alb_deinit_slave(struct bonding *bond, struct slave *slave);
void bond_alb_handle_link_change(struct bonding *bond, struct slave *slave, char link);
void bond_alb_assign_current_slave(struct bonding *bond, struct slave *new_slave);
int bond_alb_xmit(struct sk_buff *skb, struct net_device *dev);
void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave);
int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev);
void bond_alb_monitor(struct bonding *bond);
int bond_alb_set_mac_address(struct net_device *dev, void *addr);
int bond_alb_set_mac_address(struct net_device *bond_dev, void *addr);
#endif /* __BOND_ALB_H__ */
drivers/net/bonding/bond_main.c
View file @ 61b6a04a
This source diff could not be displayed because it is too large. You can view the blob instead.
drivers/net/bonding/bonding.h
View file @ 61b6a04a
......@@ -22,6 +22,9 @@
*
* 2003/05/01 - Shmulik Hen <shmulik.hen at intel dot com>
* - Added support for Transmit load balancing mode.
*
* 2003/09/24 - Shmulik Hen <shmulik.hen at intel dot com>
* - Code cleanup and style changes
*/
#ifndef _LINUX_BONDING_H
......@@ -29,152 +32,195 @@
#include <linux/timer.h>
#include <linux/proc_fs.h>
#include <linux/if_bonding.h>
#include "bond_3ad.h"
#include "bond_alb.h"
#ifdef BONDING_DEBUG
// use this like so: BOND_PRINT_DBG(("foo = %d, bar = %d", foo, bar));
#define BOND_PRINT_DBG(X) \
do { \
printk(KERN_DEBUG "%s (%d)", __FUNCTION__, __LINE__); \
printk X; \
printk("\n"); \
} while(0)
#define DRV_VERSION "2.5.0"
#define DRV_RELDATE "December 1, 2003"
#define DRV_NAME "bonding"
#define DRV_DESCRIPTION "Ethernet Channel Bonding Driver"
#ifdef BONDING_DEBUG
#define dprintk(fmt, args...) \
printk(KERN_DEBUG \
DRV_NAME ": %s() %d: " fmt, __FUNCTION__, __LINE__ , ## args )
#else
#define BOND_PRINT_DBG(X)
#define dprintk(fmt, args...)
#endif /* BONDING_DEBUG */
#define IS_UP(dev) ((((dev)->flags & (IFF_UP)) == (IFF_UP)) && \
(netif_running(dev) && netif_carrier_ok(dev)))
#define IS_UP(dev) \
((((dev)->flags & IFF_UP) == IFF_UP) && \
netif_running(dev) && \
netif_carrier_ok(dev))
/*
* Checks whether bond is ready for transmit.
*
* Caller must hold bond->lock
*/
#define BOND_IS_OK(bond) \
(((bond)->dev->flags & IFF_UP) && \
netif_running((bond)->dev) && \
((bond)->slave_cnt > 0))
/* Checks whether the dev is ready for transmit. We do not check netif_running
* since a device can be stopped by the driver for short periods of time for
* maintainance. dev_queue_xmit() handles this by queing the packet until the
* the dev is running again. Keeping packets ordering requires sticking the
* same dev as much as possible
/*
* Checks whether slave is ready for transmit.
*/
#define SLAVE_IS_OK(slave) \
((((slave)->dev->flags & (IFF_UP)) == (IFF_UP)) && \
netif_carrier_ok((slave)->dev) && \
(((slave)->dev->flags & IFF_UP) && \
netif_running((slave)->dev) && \
((slave)->link == BOND_LINK_UP) && \
((slave)->state == BOND_STATE_ACTIVE))
typedef struct slave {
#define USES_PRIMARY(mode) \
(((mode) == BOND_MODE_ACTIVEBACKUP) || \
((mode) == BOND_MODE_TLB) || \
((mode) == BOND_MODE_ALB))
/*
* Less bad way to call ioctl from within the kernel; this needs to be
* done some other way to get the call out of interrupt context.
* Needs "ioctl" variable to be supplied by calling context.
*/
#define IOCTL(dev, arg, cmd) ({ \
int res = 0; \
mm_segment_t fs = get_fs(); \
set_fs(get_ds()); \
res = ioctl(dev, arg, cmd); \
set_fs(fs); \
res; })
/**
* bond_for_each_slave_from - iterate the slaves list from a starting point
* @bond: the bond holding this list.
* @pos: current slave.
* @cnt: counter for max number of moves
* @start: starting point.
*
* Caller must hold bond->lock
*/
#define bond_for_each_slave_from(bond, pos, cnt, start) \
for (cnt = 0, pos = start; \
cnt < (bond)->slave_cnt; \
cnt++, pos = (pos)->next)
/**
* bond_for_each_slave_from_to - iterate the slaves list from start point to stop point
* @bond: the bond holding this list.
* @pos: current slave.
* @cnt: counter for number max of moves
* @start: start point.
* @stop: stop point.
*
* Caller must hold bond->lock
*/
#define bond_for_each_slave_from_to(bond, pos, cnt, start, stop) \
for (cnt = 0, pos = start; \
((cnt < (bond)->slave_cnt) && (pos != (stop)->next)); \
cnt++, pos = (pos)->next)
/**
* bond_for_each_slave - iterate the slaves list from head
* @bond: the bond holding this list.
* @pos: current slave.
* @cnt: counter for max number of moves
*
* Caller must hold bond->lock
*/
#define bond_for_each_slave(bond, pos, cnt) \
bond_for_each_slave_from(bond, pos, cnt, (bond)->first_slave)
struct slave {
struct net_device *dev; /* first - usefull for panic debug */
struct slave *next;
struct slave *prev;
struct net_device *dev;
short delay;
unsigned long jiffies;
char link; /* one of BOND_LINK_XXXX */
char state; /* one of BOND_STATE_XXXX */
unsigned short original_flags;
s16 delay;
u32 jiffies;
s8 link; /* one of BOND_LINK_XXXX */
s8 state; /* one of BOND_STATE_XXXX */
u32 original_flags;
u32 link_failure_count;
u16 speed;
u8 duplex;
u8 perm_hwaddr[ETH_ALEN];
struct ad_slave_info ad_info; /* HUGE - better to dynamically alloc */
struct tlb_slave_info tlb_info;
} slave_t;
};
/*
* Here are the locking policies for the two bonding locks:
*
* 1) Get bond->lock when reading/writing slave list.
* 2) Get bond->ptrlock when reading/writing bond->current_slave.
* 2) Get bond->curr_slave_lock when reading/writing bond->curr_active_slave.
* (It is unnecessary when the write-lock is put with bond->lock.)
* 3) When we lock with bond->ptrlock, we must lock with bond->lock
* 3) When we lock with bond->curr_slave_lock, we must lock with bond->lock
* beforehand.
*/
typedef struct bonding {
slave_t *next;
slave_t *prev;
slave_t *current_slave;
slave_t *primary_slave;
slave_t *current_arp_slave;
__s32 slave_cnt;
struct bonding {
struct net_device *dev; /* first - usefull for panic debug */
struct slave *first_slave;
struct slave *curr_active_slave;
struct slave *current_arp_slave;
struct slave *primary_slave;
s32 slave_cnt; /* never change this value outside the attach/detach wrappers */
rwlock_t lock;
rwlock_t ptrlock;
rwlock_t curr_slave_lock;
struct timer_list mii_timer;
struct timer_list arp_timer;
s8 kill_timers;
struct net_device_stats stats;
#ifdef CONFIG_PROC_FS
struct proc_dir_entry *bond_proc_file;
char procdir_name[IFNAMSIZ];
struct proc_dir_entry *proc_entry;
char proc_file_name[IFNAMSIZ];
#endif /* CONFIG_PROC_FS */
struct list_head bond_list;
struct net_device *device;
struct dev_mc_list *mc_list;
unsigned short flags;
u16 flags;
struct ad_bond_info ad_info;
struct alb_bond_info alb_info;
} bonding_t;
/* Forward declarations */
void bond_set_slave_active_flags(slave_t *slave);
void bond_set_slave_inactive_flags(slave_t *slave);
/**
* These functions can be used for iterating the slave list
* (which is circular)
* Caller must hold bond lock for read
*/
extern inline struct slave*
bond_get_first_slave(struct bonding *bond)
{
/* if there are no slaves return NULL */
if (bond->next == (slave_t *)bond) {
return NULL;
}
return bond->next;
}
/**
* Caller must hold bond lock for read
*/
extern inline struct slave*
bond_get_next_slave(struct bonding *bond, struct slave *slave)
{
/* If we have reached the last slave return NULL */
if (slave->next == bond->next) {
return NULL;
}
return slave->next;
}
};
/**
* Returns NULL if the net_device does not belong to any of the bond's slaves
*
* Caller must hold bond lock for read
*/
extern inline struct slave*
bond_get_slave_by_dev(struct bonding *bond, struct net_device *slave_dev)
extern inline struct slave *bond_get_slave_by_dev(struct bonding *bond, struct net_device *slave_dev)
{
struct slave *our_slave = bond->next;
struct slave *slave = NULL;
int i;
/* check if the list of slaves is empty */
if (our_slave == (slave_t *)bond) {
return NULL;
}
for (; our_slave; our_slave = bond_get_next_slave(bond, our_slave)) {
if (our_slave->dev == slave_dev) {
bond_for_each_slave(bond, slave, i) {
if (slave->dev == slave_dev) {
break;
}
}
return our_slave;
return slave;
}
extern inline struct bonding*
bond_get_bond_by_slave(struct slave *slave)
extern inline struct bonding *bond_get_bond_by_slave(struct slave *slave)
{
if (!slave || !slave->dev->master) {
return NULL;
}
return (struct bonding *)(slave->dev->master->priv);
return (struct bonding *)slave->dev->master->priv;
}
extern inline void bond_set_slave_inactive_flags(struct slave *slave)
{
slave->state = BOND_STATE_BACKUP;
slave->dev->flags |= IFF_NOARP;
}
extern inline void bond_set_slave_active_flags(struct slave *slave)
{
slave->state = BOND_STATE_ACTIVE;
slave->dev->flags &= ~IFF_NOARP;
}
#endif /* _LINUX_BONDING_H */
......
include/linux/if_bonding.h
View file @ 61b6a04a
......@@ -42,7 +42,7 @@
#include <linux/if_ether.h>
/* userland - kernel ABI version (2003/05/08) */
#define BOND_ABI_VERSION 1
#define BOND_ABI_VERSION 2
/*
* We can remove these ioctl definitions in 2.5. People should use the
......@@ -77,10 +77,6 @@
#define BOND_DEFAULT_MAX_BONDS 1 /* Default maximum number of devices to support */
#define BOND_MULTICAST_DISABLED 0
#define BOND_MULTICAST_ACTIVE 1
#define BOND_MULTICAST_ALL 2
typedef struct ifbond {
__s32 bond_mode;
__s32 num_slaves;
......@@ -90,9 +86,9 @@ typedef struct ifbond {
typedef struct ifslave
{
__s32 slave_id; /* Used as an IN param to the BOND_SLAVE_INFO_QUERY ioctl */
char slave_name[IFNAMSIZ];
char link;
char state;
__s8 slave_name[IFNAMSIZ];
__s8 link;
__s8 state;
__u32 link_failure_count;
} ifslave;
......@@ -115,3 +111,4 @@ struct ad_info {
* tab-width: 8
* End:
*/
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7