OpenCloudOS-Kernel/net/decnet/dn_dev.c

1451 lines
32 KiB
C
Raw Normal View History

/*
* DECnet An implementation of the DECnet protocol suite for the LINUX
* operating system. DECnet is implemented using the BSD Socket
* interface as the means of communication with the user level.
*
* DECnet Device Layer
*
* Authors: Steve Whitehouse <SteveW@ACM.org>
* Eduardo Marcelo Serrat <emserrat@geocities.com>
*
* Changes:
* Steve Whitehouse : Devices now see incoming frames so they
* can mark on who it came from.
* Steve Whitehouse : Fixed bug in creating neighbours. Each neighbour
* can now have a device specific setup func.
* Steve Whitehouse : Added /proc/sys/net/decnet/conf/<dev>/
* Steve Whitehouse : Fixed bug which sometimes killed timer
* Steve Whitehouse : Multiple ifaddr support
* Steve Whitehouse : SIOCGIFCONF is now a compile time option
* Steve Whitehouse : /proc/sys/net/decnet/conf/<sys>/forwarding
* Steve Whitehouse : Removed timer1 - it's a user space issue now
* Patrick Caulfield : Fixed router hello message format
* Steve Whitehouse : Got rid of constant sizes for blksize for
* devices. All mtu based now.
*/
#include <linux/capability.h>
#include <linux/module.h>
#include <linux/moduleparam.h>
#include <linux/init.h>
#include <linux/net.h>
#include <linux/netdevice.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/timer.h>
#include <linux/string.h>
#include <linux/if_addr.h>
#include <linux/if_arp.h>
#include <linux/if_ether.h>
#include <linux/skbuff.h>
#include <linux/sysctl.h>
#include <linux/notifier.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/jiffies.h>
#include <linux/uaccess.h>
#include <net/net_namespace.h>
#include <net/neighbour.h>
#include <net/dst.h>
#include <net/flow.h>
#include <net/fib_rules.h>
#include <net/netlink.h>
#include <net/dn.h>
#include <net/dn_dev.h>
#include <net/dn_route.h>
#include <net/dn_neigh.h>
#include <net/dn_fib.h>
#define DN_IFREQ_SIZE (sizeof(struct ifreq) - sizeof(struct sockaddr) + sizeof(struct sockaddr_dn))
static char dn_rt_all_end_mcast[ETH_ALEN] = {0xAB,0x00,0x00,0x04,0x00,0x00};
static char dn_rt_all_rt_mcast[ETH_ALEN] = {0xAB,0x00,0x00,0x03,0x00,0x00};
static char dn_hiord[ETH_ALEN] = {0xAA,0x00,0x04,0x00,0x00,0x00};
static unsigned char dn_eco_version[3] = {0x02,0x00,0x00};
extern struct neigh_table dn_neigh_table;
/*
* decnet_address is kept in network order.
*/
__le16 decnet_address = 0;
static DEFINE_SPINLOCK(dndev_lock);
static struct net_device *decnet_default_device;
[PATCH] Notifier chain update: API changes The kernel's implementation of notifier chains is unsafe. There is no protection against entries being added to or removed from a chain while the chain is in use. The issues were discussed in this thread: http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2 We noticed that notifier chains in the kernel fall into two basic usage classes: "Blocking" chains are always called from a process context and the callout routines are allowed to sleep; "Atomic" chains can be called from an atomic context and the callout routines are not allowed to sleep. We decided to codify this distinction and make it part of the API. Therefore this set of patches introduces three new, parallel APIs: one for blocking notifiers, one for atomic notifiers, and one for "raw" notifiers (which is really just the old API under a new name). New kinds of data structures are used for the heads of the chains, and new routines are defined for registration, unregistration, and calling a chain. The three APIs are explained in include/linux/notifier.h and their implementation is in kernel/sys.c. With atomic and blocking chains, the implementation guarantees that the chain links will not be corrupted and that chain callers will not get messed up by entries being added or removed. For raw chains the implementation provides no guarantees at all; users of this API must provide their own protections. (The idea was that situations may come up where the assumptions of the atomic and blocking APIs are not appropriate, so it should be possible for users to handle these things in their own way.) There are some limitations, which should not be too hard to live with. For atomic/blocking chains, registration and unregistration must always be done in a process context since the chain is protected by a mutex/rwsem. Also, a callout routine for a non-raw chain must not try to register or unregister entries on its own chain. (This did happen in a couple of places and the code had to be changed to avoid it.) Since atomic chains may be called from within an NMI handler, they cannot use spinlocks for synchronization. Instead we use RCU. The overhead falls almost entirely in the unregister routine, which is okay since unregistration is much less frequent that calling a chain. Here is the list of chains that we adjusted and their classifications. None of them use the raw API, so for the moment it is only a placeholder. ATOMIC CHAINS ------------- arch/i386/kernel/traps.c: i386die_chain arch/ia64/kernel/traps.c: ia64die_chain arch/powerpc/kernel/traps.c: powerpc_die_chain arch/sparc64/kernel/traps.c: sparc64die_chain arch/x86_64/kernel/traps.c: die_chain drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list kernel/panic.c: panic_notifier_list kernel/profile.c: task_free_notifier net/bluetooth/hci_core.c: hci_notifier net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain net/ipv6/addrconf.c: inet6addr_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain net/netlink/af_netlink.c: netlink_chain BLOCKING CHAINS --------------- arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain arch/s390/kernel/process.c: idle_chain arch/x86_64/kernel/process.c idle_notifier drivers/base/memory.c: memory_chain drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list drivers/macintosh/adb.c: adb_client_list drivers/macintosh/via-pmu.c sleep_notifier_list drivers/macintosh/via-pmu68k.c sleep_notifier_list drivers/macintosh/windfarm_core.c wf_client_list drivers/usb/core/notify.c usb_notifier_list drivers/video/fbmem.c fb_notifier_list kernel/cpu.c cpu_chain kernel/module.c module_notify_list kernel/profile.c munmap_notifier kernel/profile.c task_exit_notifier kernel/sys.c reboot_notifier_list net/core/dev.c netdev_chain net/decnet/dn_dev.c: dnaddr_chain net/ipv4/devinet.c: inetaddr_chain It's possible that some of these classifications are wrong. If they are, please let us know or submit a patch to fix them. Note that any chain that gets called very frequently should be atomic, because the rwsem read-locking used for blocking chains is very likely to incur cache misses on SMP systems. (However, if the chain's callout routines may sleep then the chain cannot be atomic.) The patch set was written by Alan Stern and Chandra Seetharaman, incorporating material written by Keith Owens and suggestions from Paul McKenney and Andrew Morton. [jes@sgi.com: restructure the notifier chain initialization macros] Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com> Signed-off-by: Jes Sorensen <jes@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-27 17:16:30 +08:00
static BLOCKING_NOTIFIER_HEAD(dnaddr_chain);
static struct dn_dev *dn_dev_create(struct net_device *dev, int *err);
static void dn_dev_delete(struct net_device *dev);
static void dn_ifaddr_notify(int event, struct dn_ifaddr *ifa);
static int dn_eth_up(struct net_device *);
static void dn_eth_down(struct net_device *);
static void dn_send_brd_hello(struct net_device *dev, struct dn_ifaddr *ifa);
static void dn_send_ptp_hello(struct net_device *dev, struct dn_ifaddr *ifa);
static struct dn_dev_parms dn_dev_list[] = {
{
.type = ARPHRD_ETHER, /* Ethernet */
.mode = DN_DEV_BCAST,
.state = DN_DEV_S_RU,
.t2 = 1,
.t3 = 10,
.name = "ethernet",
.up = dn_eth_up,
.down = dn_eth_down,
.timer3 = dn_send_brd_hello,
},
{
.type = ARPHRD_IPGRE, /* DECnet tunneled over GRE in IP */
.mode = DN_DEV_BCAST,
.state = DN_DEV_S_RU,
.t2 = 1,
.t3 = 10,
.name = "ipgre",
.timer3 = dn_send_brd_hello,
},
#if 0
{
.type = ARPHRD_X25, /* Bog standard X.25 */
.mode = DN_DEV_UCAST,
.state = DN_DEV_S_DS,
.t2 = 1,
.t3 = 120,
.name = "x25",
.timer3 = dn_send_ptp_hello,
},
#endif
#if 0
{
.type = ARPHRD_PPP, /* DECnet over PPP */
.mode = DN_DEV_BCAST,
.state = DN_DEV_S_RU,
.t2 = 1,
.t3 = 10,
.name = "ppp",
.timer3 = dn_send_brd_hello,
},
#endif
{
.type = ARPHRD_DDCMP, /* DECnet over DDCMP */
.mode = DN_DEV_UCAST,
.state = DN_DEV_S_DS,
.t2 = 1,
.t3 = 120,
.name = "ddcmp",
.timer3 = dn_send_ptp_hello,
},
{
.type = ARPHRD_LOOPBACK, /* Loopback interface - always last */
.mode = DN_DEV_BCAST,
.state = DN_DEV_S_RU,
.t2 = 1,
.t3 = 10,
.name = "loopback",
.timer3 = dn_send_brd_hello,
}
};
#define DN_DEV_LIST_SIZE ARRAY_SIZE(dn_dev_list)
#define DN_DEV_PARMS_OFFSET(x) offsetof(struct dn_dev_parms, x)
#ifdef CONFIG_SYSCTL
static int min_t2[] = { 1 };
static int max_t2[] = { 60 }; /* No max specified, but this seems sensible */
static int min_t3[] = { 1 };
static int max_t3[] = { 8191 }; /* Must fit in 16 bits when multiplied by BCT3MULT or T3MULT */
static int min_priority[1];
static int max_priority[] = { 127 }; /* From DECnet spec */
static int dn_forwarding_proc(struct ctl_table *, int,
void __user *, size_t *, loff_t *);
static struct dn_dev_sysctl_table {
struct ctl_table_header *sysctl_header;
struct ctl_table dn_dev_vars[5];
} dn_dev_sysctl = {
NULL,
{
{
.procname = "forwarding",
.data = (void *)DN_DEV_PARMS_OFFSET(forwarding),
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = dn_forwarding_proc,
},
{
.procname = "priority",
.data = (void *)DN_DEV_PARMS_OFFSET(priority),
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = &min_priority,
.extra2 = &max_priority
},
{
.procname = "t2",
.data = (void *)DN_DEV_PARMS_OFFSET(t2),
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = &min_t2,
.extra2 = &max_t2
},
{
.procname = "t3",
.data = (void *)DN_DEV_PARMS_OFFSET(t3),
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = &min_t3,
.extra2 = &max_t3
},
{ }
},
};
static void dn_dev_sysctl_register(struct net_device *dev, struct dn_dev_parms *parms)
{
struct dn_dev_sysctl_table *t;
int i;
char path[sizeof("net/decnet/conf/") + IFNAMSIZ];
t = kmemdup(&dn_dev_sysctl, sizeof(*t), GFP_KERNEL);
if (t == NULL)
return;
for(i = 0; i < ARRAY_SIZE(t->dn_dev_vars) - 1; i++) {
long offset = (long)t->dn_dev_vars[i].data;
t->dn_dev_vars[i].data = ((char *)parms) + offset;
}
snprintf(path, sizeof(path), "net/decnet/conf/%s",
dev? dev->name : parms->name);
t->dn_dev_vars[0].extra1 = (void *)dev;
t->sysctl_header = register_net_sysctl(&init_net, path, t->dn_dev_vars);
if (t->sysctl_header == NULL)
kfree(t);
else
parms->sysctl = t;
}
static void dn_dev_sysctl_unregister(struct dn_dev_parms *parms)
{
if (parms->sysctl) {
struct dn_dev_sysctl_table *t = parms->sysctl;
parms->sysctl = NULL;
unregister_net_sysctl_table(t->sysctl_header);
kfree(t);
}
}
static int dn_forwarding_proc(struct ctl_table *table, int write,
void __user *buffer,
size_t *lenp, loff_t *ppos)
{
#ifdef CONFIG_DECNET_ROUTER
struct net_device *dev = table->extra1;
struct dn_dev *dn_db;
int err;
int tmp, old;
if (table->extra1 == NULL)
return -EINVAL;
dn_db = rcu_dereference_raw(dev->dn_ptr);
old = dn_db->parms.forwarding;
err = proc_dointvec(table, write, buffer, lenp, ppos);
if ((err >= 0) && write) {
if (dn_db->parms.forwarding < 0)
dn_db->parms.forwarding = 0;
if (dn_db->parms.forwarding > 2)
dn_db->parms.forwarding = 2;
/*
* What an ugly hack this is... its works, just. It
* would be nice if sysctl/proc were just that little
* bit more flexible so I don't have to write a special
* routine, or suffer hacks like this - SJW
*/
tmp = dn_db->parms.forwarding;
dn_db->parms.forwarding = old;
if (dn_db->parms.down)
dn_db->parms.down(dev);
dn_db->parms.forwarding = tmp;
if (dn_db->parms.up)
dn_db->parms.up(dev);
}
return err;
#else
return -EINVAL;
#endif
}
#else /* CONFIG_SYSCTL */
static void dn_dev_sysctl_unregister(struct dn_dev_parms *parms)
{
}
static void dn_dev_sysctl_register(struct net_device *dev, struct dn_dev_parms *parms)
{
}
#endif /* CONFIG_SYSCTL */
static inline __u16 mtu2blksize(struct net_device *dev)
{
u32 blksize = dev->mtu;
if (blksize > 0xffff)
blksize = 0xffff;
if (dev->type == ARPHRD_ETHER ||
dev->type == ARPHRD_PPP ||
dev->type == ARPHRD_IPGRE ||
dev->type == ARPHRD_LOOPBACK)
blksize -= 2;
return (__u16)blksize;
}
static struct dn_ifaddr *dn_dev_alloc_ifa(void)
{
struct dn_ifaddr *ifa;
ifa = kzalloc(sizeof(*ifa), GFP_KERNEL);
return ifa;
}
static void dn_dev_free_ifa(struct dn_ifaddr *ifa)
{
kfree_rcu(ifa, rcu);
}
static void dn_dev_del_ifa(struct dn_dev *dn_db, struct dn_ifaddr __rcu **ifap, int destroy)
{
struct dn_ifaddr *ifa1 = rtnl_dereference(*ifap);
unsigned char mac_addr[6];
struct net_device *dev = dn_db->dev;
ASSERT_RTNL();
*ifap = ifa1->ifa_next;
if (dn_db->dev->type == ARPHRD_ETHER) {
if (ifa1->ifa_local != dn_eth2dn(dev->dev_addr)) {
dn_dn2eth(mac_addr, ifa1->ifa_local);
dev_mc_del(dev, mac_addr);
}
}
dn_ifaddr_notify(RTM_DELADDR, ifa1);
[PATCH] Notifier chain update: API changes The kernel's implementation of notifier chains is unsafe. There is no protection against entries being added to or removed from a chain while the chain is in use. The issues were discussed in this thread: http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2 We noticed that notifier chains in the kernel fall into two basic usage classes: "Blocking" chains are always called from a process context and the callout routines are allowed to sleep; "Atomic" chains can be called from an atomic context and the callout routines are not allowed to sleep. We decided to codify this distinction and make it part of the API. Therefore this set of patches introduces three new, parallel APIs: one for blocking notifiers, one for atomic notifiers, and one for "raw" notifiers (which is really just the old API under a new name). New kinds of data structures are used for the heads of the chains, and new routines are defined for registration, unregistration, and calling a chain. The three APIs are explained in include/linux/notifier.h and their implementation is in kernel/sys.c. With atomic and blocking chains, the implementation guarantees that the chain links will not be corrupted and that chain callers will not get messed up by entries being added or removed. For raw chains the implementation provides no guarantees at all; users of this API must provide their own protections. (The idea was that situations may come up where the assumptions of the atomic and blocking APIs are not appropriate, so it should be possible for users to handle these things in their own way.) There are some limitations, which should not be too hard to live with. For atomic/blocking chains, registration and unregistration must always be done in a process context since the chain is protected by a mutex/rwsem. Also, a callout routine for a non-raw chain must not try to register or unregister entries on its own chain. (This did happen in a couple of places and the code had to be changed to avoid it.) Since atomic chains may be called from within an NMI handler, they cannot use spinlocks for synchronization. Instead we use RCU. The overhead falls almost entirely in the unregister routine, which is okay since unregistration is much less frequent that calling a chain. Here is the list of chains that we adjusted and their classifications. None of them use the raw API, so for the moment it is only a placeholder. ATOMIC CHAINS ------------- arch/i386/kernel/traps.c: i386die_chain arch/ia64/kernel/traps.c: ia64die_chain arch/powerpc/kernel/traps.c: powerpc_die_chain arch/sparc64/kernel/traps.c: sparc64die_chain arch/x86_64/kernel/traps.c: die_chain drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list kernel/panic.c: panic_notifier_list kernel/profile.c: task_free_notifier net/bluetooth/hci_core.c: hci_notifier net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain net/ipv6/addrconf.c: inet6addr_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain net/netlink/af_netlink.c: netlink_chain BLOCKING CHAINS --------------- arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain arch/s390/kernel/process.c: idle_chain arch/x86_64/kernel/process.c idle_notifier drivers/base/memory.c: memory_chain drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list drivers/macintosh/adb.c: adb_client_list drivers/macintosh/via-pmu.c sleep_notifier_list drivers/macintosh/via-pmu68k.c sleep_notifier_list drivers/macintosh/windfarm_core.c wf_client_list drivers/usb/core/notify.c usb_notifier_list drivers/video/fbmem.c fb_notifier_list kernel/cpu.c cpu_chain kernel/module.c module_notify_list kernel/profile.c munmap_notifier kernel/profile.c task_exit_notifier kernel/sys.c reboot_notifier_list net/core/dev.c netdev_chain net/decnet/dn_dev.c: dnaddr_chain net/ipv4/devinet.c: inetaddr_chain It's possible that some of these classifications are wrong. If they are, please let us know or submit a patch to fix them. Note that any chain that gets called very frequently should be atomic, because the rwsem read-locking used for blocking chains is very likely to incur cache misses on SMP systems. (However, if the chain's callout routines may sleep then the chain cannot be atomic.) The patch set was written by Alan Stern and Chandra Seetharaman, incorporating material written by Keith Owens and suggestions from Paul McKenney and Andrew Morton. [jes@sgi.com: restructure the notifier chain initialization macros] Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com> Signed-off-by: Jes Sorensen <jes@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-27 17:16:30 +08:00
blocking_notifier_call_chain(&dnaddr_chain, NETDEV_DOWN, ifa1);
if (destroy) {
dn_dev_free_ifa(ifa1);
if (dn_db->ifa_list == NULL)
dn_dev_delete(dn_db->dev);
}
}
static int dn_dev_insert_ifa(struct dn_dev *dn_db, struct dn_ifaddr *ifa)
{
struct net_device *dev = dn_db->dev;
struct dn_ifaddr *ifa1;
unsigned char mac_addr[6];
ASSERT_RTNL();
/* Check for duplicates */
for (ifa1 = rtnl_dereference(dn_db->ifa_list);
ifa1 != NULL;
ifa1 = rtnl_dereference(ifa1->ifa_next)) {
if (ifa1->ifa_local == ifa->ifa_local)
return -EEXIST;
}
if (dev->type == ARPHRD_ETHER) {
if (ifa->ifa_local != dn_eth2dn(dev->dev_addr)) {
dn_dn2eth(mac_addr, ifa->ifa_local);
dev_mc_add(dev, mac_addr);
}
}
ifa->ifa_next = dn_db->ifa_list;
rcu_assign_pointer(dn_db->ifa_list, ifa);
dn_ifaddr_notify(RTM_NEWADDR, ifa);
[PATCH] Notifier chain update: API changes The kernel's implementation of notifier chains is unsafe. There is no protection against entries being added to or removed from a chain while the chain is in use. The issues were discussed in this thread: http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2 We noticed that notifier chains in the kernel fall into two basic usage classes: "Blocking" chains are always called from a process context and the callout routines are allowed to sleep; "Atomic" chains can be called from an atomic context and the callout routines are not allowed to sleep. We decided to codify this distinction and make it part of the API. Therefore this set of patches introduces three new, parallel APIs: one for blocking notifiers, one for atomic notifiers, and one for "raw" notifiers (which is really just the old API under a new name). New kinds of data structures are used for the heads of the chains, and new routines are defined for registration, unregistration, and calling a chain. The three APIs are explained in include/linux/notifier.h and their implementation is in kernel/sys.c. With atomic and blocking chains, the implementation guarantees that the chain links will not be corrupted and that chain callers will not get messed up by entries being added or removed. For raw chains the implementation provides no guarantees at all; users of this API must provide their own protections. (The idea was that situations may come up where the assumptions of the atomic and blocking APIs are not appropriate, so it should be possible for users to handle these things in their own way.) There are some limitations, which should not be too hard to live with. For atomic/blocking chains, registration and unregistration must always be done in a process context since the chain is protected by a mutex/rwsem. Also, a callout routine for a non-raw chain must not try to register or unregister entries on its own chain. (This did happen in a couple of places and the code had to be changed to avoid it.) Since atomic chains may be called from within an NMI handler, they cannot use spinlocks for synchronization. Instead we use RCU. The overhead falls almost entirely in the unregister routine, which is okay since unregistration is much less frequent that calling a chain. Here is the list of chains that we adjusted and their classifications. None of them use the raw API, so for the moment it is only a placeholder. ATOMIC CHAINS ------------- arch/i386/kernel/traps.c: i386die_chain arch/ia64/kernel/traps.c: ia64die_chain arch/powerpc/kernel/traps.c: powerpc_die_chain arch/sparc64/kernel/traps.c: sparc64die_chain arch/x86_64/kernel/traps.c: die_chain drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list kernel/panic.c: panic_notifier_list kernel/profile.c: task_free_notifier net/bluetooth/hci_core.c: hci_notifier net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain net/ipv6/addrconf.c: inet6addr_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain net/netlink/af_netlink.c: netlink_chain BLOCKING CHAINS --------------- arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain arch/s390/kernel/process.c: idle_chain arch/x86_64/kernel/process.c idle_notifier drivers/base/memory.c: memory_chain drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list drivers/macintosh/adb.c: adb_client_list drivers/macintosh/via-pmu.c sleep_notifier_list drivers/macintosh/via-pmu68k.c sleep_notifier_list drivers/macintosh/windfarm_core.c wf_client_list drivers/usb/core/notify.c usb_notifier_list drivers/video/fbmem.c fb_notifier_list kernel/cpu.c cpu_chain kernel/module.c module_notify_list kernel/profile.c munmap_notifier kernel/profile.c task_exit_notifier kernel/sys.c reboot_notifier_list net/core/dev.c netdev_chain net/decnet/dn_dev.c: dnaddr_chain net/ipv4/devinet.c: inetaddr_chain It's possible that some of these classifications are wrong. If they are, please let us know or submit a patch to fix them. Note that any chain that gets called very frequently should be atomic, because the rwsem read-locking used for blocking chains is very likely to incur cache misses on SMP systems. (However, if the chain's callout routines may sleep then the chain cannot be atomic.) The patch set was written by Alan Stern and Chandra Seetharaman, incorporating material written by Keith Owens and suggestions from Paul McKenney and Andrew Morton. [jes@sgi.com: restructure the notifier chain initialization macros] Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com> Signed-off-by: Jes Sorensen <jes@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-27 17:16:30 +08:00
blocking_notifier_call_chain(&dnaddr_chain, NETDEV_UP, ifa);
return 0;
}
static int dn_dev_set_ifa(struct net_device *dev, struct dn_ifaddr *ifa)
{
struct dn_dev *dn_db = rtnl_dereference(dev->dn_ptr);
int rv;
if (dn_db == NULL) {
int err;
dn_db = dn_dev_create(dev, &err);
if (dn_db == NULL)
return err;
}
ifa->ifa_dev = dn_db;
if (dev->flags & IFF_LOOPBACK)
ifa->ifa_scope = RT_SCOPE_HOST;
rv = dn_dev_insert_ifa(dn_db, ifa);
if (rv)
dn_dev_free_ifa(ifa);
return rv;
}
int dn_dev_ioctl(unsigned int cmd, void __user *arg)
{
char buffer[DN_IFREQ_SIZE];
struct ifreq *ifr = (struct ifreq *)buffer;
struct sockaddr_dn *sdn = (struct sockaddr_dn *)&ifr->ifr_addr;
struct dn_dev *dn_db;
struct net_device *dev;
struct dn_ifaddr *ifa = NULL;
struct dn_ifaddr __rcu **ifap = NULL;
int ret = 0;
if (copy_from_user(ifr, arg, DN_IFREQ_SIZE))
return -EFAULT;
ifr->ifr_name[IFNAMSIZ-1] = 0;
[NET]: Make the device list and device lookups per namespace. This patch makes most of the generic device layer network namespace safe. This patch makes dev_base_head a network namespace variable, and then it picks up a few associated variables. The functions: dev_getbyhwaddr dev_getfirsthwbytype dev_get_by_flags dev_get_by_name __dev_get_by_name dev_get_by_index __dev_get_by_index dev_ioctl dev_ethtool dev_load wireless_process_ioctl were modified to take a network namespace argument, and deal with it. vlan_ioctl_set and brioctl_set were modified so their hooks will receive a network namespace argument. So basically anthing in the core of the network stack that was affected to by the change of dev_base was modified to handle multiple network namespaces. The rest of the network stack was simply modified to explicitly use &init_net the initial network namespace. This can be fixed when those components of the network stack are modified to handle multiple network namespaces. For now the ifindex generator is left global. Fundametally ifindex numbers are per namespace, or else we will have corner case problems with migration when we get that far. At the same time there are assumptions in the network stack that the ifindex of a network device won't change. Making the ifindex number global seems a good compromise until the network stack can cope with ifindex changes when you change namespaces, and the like. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-18 02:56:21 +08:00
dev_load(&init_net, ifr->ifr_name);
switch (cmd) {
case SIOCGIFADDR:
break;
case SIOCSIFADDR:
if (!capable(CAP_NET_ADMIN))
return -EACCES;
if (sdn->sdn_family != AF_DECnet)
return -EINVAL;
break;
default:
return -EINVAL;
}
rtnl_lock();
[NET]: Make the device list and device lookups per namespace. This patch makes most of the generic device layer network namespace safe. This patch makes dev_base_head a network namespace variable, and then it picks up a few associated variables. The functions: dev_getbyhwaddr dev_getfirsthwbytype dev_get_by_flags dev_get_by_name __dev_get_by_name dev_get_by_index __dev_get_by_index dev_ioctl dev_ethtool dev_load wireless_process_ioctl were modified to take a network namespace argument, and deal with it. vlan_ioctl_set and brioctl_set were modified so their hooks will receive a network namespace argument. So basically anthing in the core of the network stack that was affected to by the change of dev_base was modified to handle multiple network namespaces. The rest of the network stack was simply modified to explicitly use &init_net the initial network namespace. This can be fixed when those components of the network stack are modified to handle multiple network namespaces. For now the ifindex generator is left global. Fundametally ifindex numbers are per namespace, or else we will have corner case problems with migration when we get that far. At the same time there are assumptions in the network stack that the ifindex of a network device won't change. Making the ifindex number global seems a good compromise until the network stack can cope with ifindex changes when you change namespaces, and the like. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-18 02:56:21 +08:00
if ((dev = __dev_get_by_name(&init_net, ifr->ifr_name)) == NULL) {
ret = -ENODEV;
goto done;
}
if ((dn_db = rtnl_dereference(dev->dn_ptr)) != NULL) {
for (ifap = &dn_db->ifa_list;
(ifa = rtnl_dereference(*ifap)) != NULL;
ifap = &ifa->ifa_next)
if (strcmp(ifr->ifr_name, ifa->ifa_label) == 0)
break;
}
if (ifa == NULL && cmd != SIOCSIFADDR) {
ret = -EADDRNOTAVAIL;
goto done;
}
switch (cmd) {
case SIOCGIFADDR:
*((__le16 *)sdn->sdn_nodeaddr) = ifa->ifa_local;
goto rarok;
case SIOCSIFADDR:
if (!ifa) {
if ((ifa = dn_dev_alloc_ifa()) == NULL) {
ret = -ENOBUFS;
break;
}
memcpy(ifa->ifa_label, dev->name, IFNAMSIZ);
} else {
if (ifa->ifa_local == dn_saddr2dn(sdn))
break;
dn_dev_del_ifa(dn_db, ifap, 0);
}
ifa->ifa_local = ifa->ifa_address = dn_saddr2dn(sdn);
ret = dn_dev_set_ifa(dev, ifa);
}
done:
rtnl_unlock();
return ret;
rarok:
if (copy_to_user(arg, ifr, DN_IFREQ_SIZE))
ret = -EFAULT;
goto done;
}
struct net_device *dn_dev_get_default(void)
{
struct net_device *dev;
spin_lock(&dndev_lock);
dev = decnet_default_device;
if (dev) {
if (dev->dn_ptr)
dev_hold(dev);
else
dev = NULL;
}
spin_unlock(&dndev_lock);
return dev;
}
int dn_dev_set_default(struct net_device *dev, int force)
{
struct net_device *old = NULL;
int rv = -EBUSY;
if (!dev->dn_ptr)
return -ENODEV;
spin_lock(&dndev_lock);
if (force || decnet_default_device == NULL) {
old = decnet_default_device;
decnet_default_device = dev;
rv = 0;
}
spin_unlock(&dndev_lock);
if (old)
dev_put(old);
return rv;
}
static void dn_dev_check_default(struct net_device *dev)
{
spin_lock(&dndev_lock);
if (dev == decnet_default_device) {
decnet_default_device = NULL;
} else {
dev = NULL;
}
spin_unlock(&dndev_lock);
if (dev)
dev_put(dev);
}
/*
* Called with RTNL
*/
static struct dn_dev *dn_dev_by_index(int ifindex)
{
struct net_device *dev;
struct dn_dev *dn_dev = NULL;
dev = __dev_get_by_index(&init_net, ifindex);
if (dev)
dn_dev = rtnl_dereference(dev->dn_ptr);
return dn_dev;
}
static const struct nla_policy dn_ifa_policy[IFA_MAX+1] = {
[IFA_ADDRESS] = { .type = NLA_U16 },
[IFA_LOCAL] = { .type = NLA_U16 },
[IFA_LABEL] = { .type = NLA_STRING,
.len = IFNAMSIZ - 1 },
[IFA_FLAGS] = { .type = NLA_U32 },
};
static int dn_nl_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh,
struct netlink_ext_ack *extack)
{
struct net *net = sock_net(skb->sk);
struct nlattr *tb[IFA_MAX+1];
struct dn_dev *dn_db;
struct ifaddrmsg *ifm;
struct dn_ifaddr *ifa;
struct dn_ifaddr __rcu **ifap;
int err = -EINVAL;
if (!netlink_capable(skb, CAP_NET_ADMIN))
return -EPERM;
if (!net_eq(net, &init_net))
goto errout;
err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFA_MAX, dn_ifa_policy,
extack);
if (err < 0)
goto errout;
err = -ENODEV;
ifm = nlmsg_data(nlh);
if ((dn_db = dn_dev_by_index(ifm->ifa_index)) == NULL)
goto errout;
err = -EADDRNOTAVAIL;
for (ifap = &dn_db->ifa_list;
(ifa = rtnl_dereference(*ifap)) != NULL;
ifap = &ifa->ifa_next) {
if (tb[IFA_LOCAL] &&
nla_memcmp(tb[IFA_LOCAL], &ifa->ifa_local, 2))
continue;
if (tb[IFA_LABEL] && nla_strcmp(tb[IFA_LABEL], ifa->ifa_label))
continue;
dn_dev_del_ifa(dn_db, ifap, 1);
return 0;
}
errout:
return err;
}
static int dn_nl_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh,
struct netlink_ext_ack *extack)
{
struct net *net = sock_net(skb->sk);
struct nlattr *tb[IFA_MAX+1];
struct net_device *dev;
struct dn_dev *dn_db;
struct ifaddrmsg *ifm;
struct dn_ifaddr *ifa;
int err;
if (!netlink_capable(skb, CAP_NET_ADMIN))
return -EPERM;
if (!net_eq(net, &init_net))
return -EINVAL;
err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFA_MAX, dn_ifa_policy,
extack);
if (err < 0)
return err;
if (tb[IFA_LOCAL] == NULL)
return -EINVAL;
ifm = nlmsg_data(nlh);
[NET]: Make the device list and device lookups per namespace. This patch makes most of the generic device layer network namespace safe. This patch makes dev_base_head a network namespace variable, and then it picks up a few associated variables. The functions: dev_getbyhwaddr dev_getfirsthwbytype dev_get_by_flags dev_get_by_name __dev_get_by_name dev_get_by_index __dev_get_by_index dev_ioctl dev_ethtool dev_load wireless_process_ioctl were modified to take a network namespace argument, and deal with it. vlan_ioctl_set and brioctl_set were modified so their hooks will receive a network namespace argument. So basically anthing in the core of the network stack that was affected to by the change of dev_base was modified to handle multiple network namespaces. The rest of the network stack was simply modified to explicitly use &init_net the initial network namespace. This can be fixed when those components of the network stack are modified to handle multiple network namespaces. For now the ifindex generator is left global. Fundametally ifindex numbers are per namespace, or else we will have corner case problems with migration when we get that far. At the same time there are assumptions in the network stack that the ifindex of a network device won't change. Making the ifindex number global seems a good compromise until the network stack can cope with ifindex changes when you change namespaces, and the like. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-18 02:56:21 +08:00
if ((dev = __dev_get_by_index(&init_net, ifm->ifa_index)) == NULL)
return -ENODEV;
if ((dn_db = rtnl_dereference(dev->dn_ptr)) == NULL) {
dn_db = dn_dev_create(dev, &err);
if (!dn_db)
return err;
}
if ((ifa = dn_dev_alloc_ifa()) == NULL)
return -ENOBUFS;
if (tb[IFA_ADDRESS] == NULL)
tb[IFA_ADDRESS] = tb[IFA_LOCAL];
ifa->ifa_local = nla_get_le16(tb[IFA_LOCAL]);
ifa->ifa_address = nla_get_le16(tb[IFA_ADDRESS]);
ifa->ifa_flags = tb[IFA_FLAGS] ? nla_get_u32(tb[IFA_FLAGS]) :
ifm->ifa_flags;
ifa->ifa_scope = ifm->ifa_scope;
ifa->ifa_dev = dn_db;
if (tb[IFA_LABEL])
nla_strlcpy(ifa->ifa_label, tb[IFA_LABEL], IFNAMSIZ);
else
memcpy(ifa->ifa_label, dev->name, IFNAMSIZ);
err = dn_dev_insert_ifa(dn_db, ifa);
if (err)
dn_dev_free_ifa(ifa);
return err;
}
static inline size_t dn_ifaddr_nlmsg_size(void)
{
return NLMSG_ALIGN(sizeof(struct ifaddrmsg))
+ nla_total_size(IFNAMSIZ) /* IFA_LABEL */
+ nla_total_size(2) /* IFA_ADDRESS */
+ nla_total_size(2) /* IFA_LOCAL */
+ nla_total_size(4); /* IFA_FLAGS */
}
static int dn_nl_fill_ifaddr(struct sk_buff *skb, struct dn_ifaddr *ifa,
u32 portid, u32 seq, int event, unsigned int flags)
{
struct ifaddrmsg *ifm;
struct nlmsghdr *nlh;
u32 ifa_flags = ifa->ifa_flags | IFA_F_PERMANENT;
nlh = nlmsg_put(skb, portid, seq, event, sizeof(*ifm), flags);
if (nlh == NULL)
return -EMSGSIZE;
ifm = nlmsg_data(nlh);
ifm->ifa_family = AF_DECnet;
ifm->ifa_prefixlen = 16;
ifm->ifa_flags = ifa_flags;
ifm->ifa_scope = ifa->ifa_scope;
ifm->ifa_index = ifa->ifa_dev->dev->ifindex;
if ((ifa->ifa_address &&
nla_put_le16(skb, IFA_ADDRESS, ifa->ifa_address)) ||
(ifa->ifa_local &&
nla_put_le16(skb, IFA_LOCAL, ifa->ifa_local)) ||
(ifa->ifa_label[0] &&
nla_put_string(skb, IFA_LABEL, ifa->ifa_label)) ||
nla_put_u32(skb, IFA_FLAGS, ifa_flags))
goto nla_put_failure;
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-17 05:09:00 +08:00
nlmsg_end(skb, nlh);
return 0;
nla_put_failure:
nlmsg_cancel(skb, nlh);
return -EMSGSIZE;
}
static void dn_ifaddr_notify(int event, struct dn_ifaddr *ifa)
{
struct sk_buff *skb;
int err = -ENOBUFS;
skb = alloc_skb(dn_ifaddr_nlmsg_size(), GFP_KERNEL);
if (skb == NULL)
goto errout;
err = dn_nl_fill_ifaddr(skb, ifa, 0, 0, event, 0);
if (err < 0) {
/* -EMSGSIZE implies BUG in dn_ifaddr_nlmsg_size() */
WARN_ON(err == -EMSGSIZE);
kfree_skb(skb);
goto errout;
}
2009-02-25 15:18:28 +08:00
rtnl_notify(skb, &init_net, 0, RTNLGRP_DECnet_IFADDR, NULL, GFP_KERNEL);
return;
errout:
if (err < 0)
rtnl_set_sk_err(&init_net, RTNLGRP_DECnet_IFADDR, err);
}
static int dn_nl_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb)
{
struct net *net = sock_net(skb->sk);
int idx, dn_idx = 0, skip_ndevs, skip_naddr;
struct net_device *dev;
struct dn_dev *dn_db;
struct dn_ifaddr *ifa;
if (!net_eq(net, &init_net))
return 0;
skip_ndevs = cb->args[0];
skip_naddr = cb->args[1];
idx = 0;
rcu_read_lock();
for_each_netdev_rcu(&init_net, dev) {
if (idx < skip_ndevs)
goto cont;
else if (idx > skip_ndevs) {
/* Only skip over addresses for first dev dumped
* in this iteration (idx == skip_ndevs) */
skip_naddr = 0;
}
if ((dn_db = rcu_dereference(dev->dn_ptr)) == NULL)
goto cont;
for (ifa = rcu_dereference(dn_db->ifa_list), dn_idx = 0; ifa;
ifa = rcu_dereference(ifa->ifa_next), dn_idx++) {
if (dn_idx < skip_naddr)
continue;
if (dn_nl_fill_ifaddr(skb, ifa, NETLINK_CB(cb->skb).portid,
cb->nlh->nlmsg_seq, RTM_NEWADDR,
NLM_F_MULTI) < 0)
goto done;
}
cont:
idx++;
}
done:
rcu_read_unlock();
cb->args[0] = idx;
cb->args[1] = dn_idx;
return skb->len;
}
static int dn_dev_get_first(struct net_device *dev, __le16 *addr)
{
struct dn_dev *dn_db;
struct dn_ifaddr *ifa;
int rv = -ENODEV;
rcu_read_lock();
dn_db = rcu_dereference(dev->dn_ptr);
if (dn_db == NULL)
goto out;
ifa = rcu_dereference(dn_db->ifa_list);
if (ifa != NULL) {
*addr = ifa->ifa_local;
rv = 0;
}
out:
rcu_read_unlock();
return rv;
}
/*
* Find a default address to bind to.
*
* This is one of those areas where the initial VMS concepts don't really
* map onto the Linux concepts, and since we introduced multiple addresses
* per interface we have to cope with slightly odd ways of finding out what
* "our address" really is. Mostly it's not a problem; for this we just guess
* a sensible default. Eventually the routing code will take care of all the
* nasties for us I hope.
*/
int dn_dev_bind_default(__le16 *addr)
{
struct net_device *dev;
int rv;
dev = dn_dev_get_default();
last_chance:
if (dev) {
rv = dn_dev_get_first(dev, addr);
dev_put(dev);
if (rv == 0 || dev == init_net.loopback_dev)
return rv;
}
dev = init_net.loopback_dev;
dev_hold(dev);
goto last_chance;
}
static void dn_send_endnode_hello(struct net_device *dev, struct dn_ifaddr *ifa)
{
struct endnode_hello_message *msg;
struct sk_buff *skb = NULL;
__le16 *pktlen;
struct dn_dev *dn_db = rcu_dereference_raw(dev->dn_ptr);
if ((skb = dn_alloc_skb(NULL, sizeof(*msg), GFP_ATOMIC)) == NULL)
return;
skb->dev = dev;
msg = skb_put(skb, sizeof(*msg));
msg->msgflg = 0x0D;
memcpy(msg->tiver, dn_eco_version, 3);
dn_dn2eth(msg->id, ifa->ifa_local);
msg->iinfo = DN_RT_INFO_ENDN;
msg->blksize = cpu_to_le16(mtu2blksize(dev));
msg->area = 0x00;
memset(msg->seed, 0, 8);
memcpy(msg->neighbor, dn_hiord, ETH_ALEN);
if (dn_db->router) {
struct dn_neigh *dn = (struct dn_neigh *)dn_db->router;
dn_dn2eth(msg->neighbor, dn->addr);
}
msg->timer = cpu_to_le16((unsigned short)dn_db->parms.t3);
msg->mpd = 0x00;
msg->datalen = 0x02;
memset(msg->data, 0xAA, 2);
pktlen = (__le16 *)skb_push(skb,2);
*pktlen = cpu_to_le16(skb->len - 2);
skb_reset_network_header(skb);
dn_rt_finish_output(skb, dn_rt_all_rt_mcast, msg->id);
}
#define DRDELAY (5 * HZ)
static int dn_am_i_a_router(struct dn_neigh *dn, struct dn_dev *dn_db, struct dn_ifaddr *ifa)
{
/* First check time since device went up */
if (time_before(jiffies, dn_db->uptime + DRDELAY))
return 0;
/* If there is no router, then yes... */
if (!dn_db->router)
return 1;
/* otherwise only if we have a higher priority or.. */
if (dn->priority < dn_db->parms.priority)
return 1;
/* if we have equal priority and a higher node number */
if (dn->priority != dn_db->parms.priority)
return 0;
if (le16_to_cpu(dn->addr) < le16_to_cpu(ifa->ifa_local))
return 1;
return 0;
}
static void dn_send_router_hello(struct net_device *dev, struct dn_ifaddr *ifa)
{
int n;
struct dn_dev *dn_db = rcu_dereference_raw(dev->dn_ptr);
struct dn_neigh *dn = (struct dn_neigh *)dn_db->router;
struct sk_buff *skb;
size_t size;
unsigned char *ptr;
unsigned char *i1, *i2;
__le16 *pktlen;
char *src;
if (mtu2blksize(dev) < (26 + 7))
return;
n = mtu2blksize(dev) - 26;
n /= 7;
if (n > 32)
n = 32;
size = 2 + 26 + 7 * n;
if ((skb = dn_alloc_skb(NULL, size, GFP_ATOMIC)) == NULL)
return;
skb->dev = dev;
ptr = skb_put(skb, size);
*ptr++ = DN_RT_PKT_CNTL | DN_RT_PKT_ERTH;
*ptr++ = 2; /* ECO */
*ptr++ = 0;
*ptr++ = 0;
dn_dn2eth(ptr, ifa->ifa_local);
src = ptr;
ptr += ETH_ALEN;
*ptr++ = dn_db->parms.forwarding == 1 ?
DN_RT_INFO_L1RT : DN_RT_INFO_L2RT;
*((__le16 *)ptr) = cpu_to_le16(mtu2blksize(dev));
ptr += 2;
*ptr++ = dn_db->parms.priority; /* Priority */
*ptr++ = 0; /* Area: Reserved */
*((__le16 *)ptr) = cpu_to_le16((unsigned short)dn_db->parms.t3);
ptr += 2;
*ptr++ = 0; /* MPD: Reserved */
i1 = ptr++;
memset(ptr, 0, 7); /* Name: Reserved */
ptr += 7;
i2 = ptr++;
n = dn_neigh_elist(dev, ptr, n);
*i2 = 7 * n;
*i1 = 8 + *i2;
skb_trim(skb, (27 + *i2));
pktlen = (__le16 *)skb_push(skb, 2);
*pktlen = cpu_to_le16(skb->len - 2);
skb_reset_network_header(skb);
if (dn_am_i_a_router(dn, dn_db, ifa)) {
struct sk_buff *skb2 = skb_copy(skb, GFP_ATOMIC);
if (skb2) {
dn_rt_finish_output(skb2, dn_rt_all_end_mcast, src);
}
}
dn_rt_finish_output(skb, dn_rt_all_rt_mcast, src);
}
static void dn_send_brd_hello(struct net_device *dev, struct dn_ifaddr *ifa)
{
struct dn_dev *dn_db = rcu_dereference_raw(dev->dn_ptr);
if (dn_db->parms.forwarding == 0)
dn_send_endnode_hello(dev, ifa);
else
dn_send_router_hello(dev, ifa);
}
static void dn_send_ptp_hello(struct net_device *dev, struct dn_ifaddr *ifa)
{
int tdlen = 16;
int size = dev->hard_header_len + 2 + 4 + tdlen;
struct sk_buff *skb = dn_alloc_skb(NULL, size, GFP_ATOMIC);
int i;
unsigned char *ptr;
char src[ETH_ALEN];
if (skb == NULL)
return ;
skb->dev = dev;
skb_push(skb, dev->hard_header_len);
ptr = skb_put(skb, 2 + 4 + tdlen);
*ptr++ = DN_RT_PKT_HELO;
*((__le16 *)ptr) = ifa->ifa_local;
ptr += 2;
*ptr++ = tdlen;
for(i = 0; i < tdlen; i++)
*ptr++ = 0252;
dn_dn2eth(src, ifa->ifa_local);
dn_rt_finish_output(skb, dn_rt_all_rt_mcast, src);
}
static int dn_eth_up(struct net_device *dev)
{
struct dn_dev *dn_db = rcu_dereference_raw(dev->dn_ptr);
if (dn_db->parms.forwarding == 0)
dev_mc_add(dev, dn_rt_all_end_mcast);
else
dev_mc_add(dev, dn_rt_all_rt_mcast);
dn_db->use_long = 1;
return 0;
}
static void dn_eth_down(struct net_device *dev)
{
struct dn_dev *dn_db = rcu_dereference_raw(dev->dn_ptr);
if (dn_db->parms.forwarding == 0)
dev_mc_del(dev, dn_rt_all_end_mcast);
else
dev_mc_del(dev, dn_rt_all_rt_mcast);
}
static void dn_dev_set_timer(struct net_device *dev);
static void dn_dev_timer_func(unsigned long arg)
{
struct net_device *dev = (struct net_device *)arg;
struct dn_dev *dn_db;
struct dn_ifaddr *ifa;
rcu_read_lock();
dn_db = rcu_dereference(dev->dn_ptr);
if (dn_db->t3 <= dn_db->parms.t2) {
if (dn_db->parms.timer3) {
for (ifa = rcu_dereference(dn_db->ifa_list);
ifa;
ifa = rcu_dereference(ifa->ifa_next)) {
if (!(ifa->ifa_flags & IFA_F_SECONDARY))
dn_db->parms.timer3(dev, ifa);
}
}
dn_db->t3 = dn_db->parms.t3;
} else {
dn_db->t3 -= dn_db->parms.t2;
}
rcu_read_unlock();
dn_dev_set_timer(dev);
}
static void dn_dev_set_timer(struct net_device *dev)
{
struct dn_dev *dn_db = rcu_dereference_raw(dev->dn_ptr);
if (dn_db->parms.t2 > dn_db->parms.t3)
dn_db->parms.t2 = dn_db->parms.t3;
dn_db->timer.data = (unsigned long)dev;
dn_db->timer.function = dn_dev_timer_func;
dn_db->timer.expires = jiffies + (dn_db->parms.t2 * HZ);
add_timer(&dn_db->timer);
}
static struct dn_dev *dn_dev_create(struct net_device *dev, int *err)
{
int i;
struct dn_dev_parms *p = dn_dev_list;
struct dn_dev *dn_db;
for(i = 0; i < DN_DEV_LIST_SIZE; i++, p++) {
if (p->type == dev->type)
break;
}
*err = -ENODEV;
if (i == DN_DEV_LIST_SIZE)
return NULL;
*err = -ENOBUFS;
if ((dn_db = kzalloc(sizeof(struct dn_dev), GFP_ATOMIC)) == NULL)
return NULL;
memcpy(&dn_db->parms, p, sizeof(struct dn_dev_parms));
rcu_assign_pointer(dev->dn_ptr, dn_db);
dn_db->dev = dev;
init_timer(&dn_db->timer);
dn_db->uptime = jiffies;
dn_db->neigh_parms = neigh_parms_alloc(dev, &dn_neigh_table);
if (!dn_db->neigh_parms) {
RCU_INIT_POINTER(dev->dn_ptr, NULL);
kfree(dn_db);
return NULL;
}
if (dn_db->parms.up) {
if (dn_db->parms.up(dev) < 0) {
neigh_parms_release(&dn_neigh_table, dn_db->neigh_parms);
dev->dn_ptr = NULL;
kfree(dn_db);
return NULL;
}
}
dn_dev_sysctl_register(dev, &dn_db->parms);
dn_dev_set_timer(dev);
*err = 0;
return dn_db;
}
/*
* This processes a device up event. We only start up
* the loopback device & ethernet devices with correct
* MAC addresses automatically. Others must be started
* specifically.
*
* FIXME: How should we configure the loopback address ? If we could dispense
* with using decnet_address here and for autobind, it will be one less thing
* for users to worry about setting up.
*/
void dn_dev_up(struct net_device *dev)
{
struct dn_ifaddr *ifa;
__le16 addr = decnet_address;
int maybe_default = 0;
struct dn_dev *dn_db = rtnl_dereference(dev->dn_ptr);
if ((dev->type != ARPHRD_ETHER) && (dev->type != ARPHRD_LOOPBACK))
return;
/*
* Need to ensure that loopback device has a dn_db attached to it
* to allow creation of neighbours against it, even though it might
* not have a local address of its own. Might as well do the same for
* all autoconfigured interfaces.
*/
if (dn_db == NULL) {
int err;
dn_db = dn_dev_create(dev, &err);
if (dn_db == NULL)
return;
}
if (dev->type == ARPHRD_ETHER) {
if (memcmp(dev->dev_addr, dn_hiord, 4) != 0)
return;
addr = dn_eth2dn(dev->dev_addr);
maybe_default = 1;
}
if (addr == 0)
return;
if ((ifa = dn_dev_alloc_ifa()) == NULL)
return;
ifa->ifa_local = ifa->ifa_address = addr;
ifa->ifa_flags = 0;
ifa->ifa_scope = RT_SCOPE_UNIVERSE;
strcpy(ifa->ifa_label, dev->name);
dn_dev_set_ifa(dev, ifa);
/*
* Automagically set the default device to the first automatically
* configured ethernet card in the system.
*/
if (maybe_default) {
dev_hold(dev);
if (dn_dev_set_default(dev, 0))
dev_put(dev);
}
}
static void dn_dev_delete(struct net_device *dev)
{
struct dn_dev *dn_db = rtnl_dereference(dev->dn_ptr);
if (dn_db == NULL)
return;
del_timer_sync(&dn_db->timer);
dn_dev_sysctl_unregister(&dn_db->parms);
dn_dev_check_default(dev);
neigh_ifdown(&dn_neigh_table, dev);
if (dn_db->parms.down)
dn_db->parms.down(dev);
dev->dn_ptr = NULL;
neigh_parms_release(&dn_neigh_table, dn_db->neigh_parms);
neigh_ifdown(&dn_neigh_table, dev);
if (dn_db->router)
neigh_release(dn_db->router);
if (dn_db->peer)
neigh_release(dn_db->peer);
kfree(dn_db);
}
void dn_dev_down(struct net_device *dev)
{
struct dn_dev *dn_db = rtnl_dereference(dev->dn_ptr);
struct dn_ifaddr *ifa;
if (dn_db == NULL)
return;
while ((ifa = rtnl_dereference(dn_db->ifa_list)) != NULL) {
dn_dev_del_ifa(dn_db, &dn_db->ifa_list, 0);
dn_dev_free_ifa(ifa);
}
dn_dev_delete(dev);
}
void dn_dev_init_pkt(struct sk_buff *skb)
{
}
void dn_dev_veri_pkt(struct sk_buff *skb)
{
}
void dn_dev_hello(struct sk_buff *skb)
{
}
void dn_dev_devices_off(void)
{
struct net_device *dev;
rtnl_lock();
[NET]: Make the device list and device lookups per namespace. This patch makes most of the generic device layer network namespace safe. This patch makes dev_base_head a network namespace variable, and then it picks up a few associated variables. The functions: dev_getbyhwaddr dev_getfirsthwbytype dev_get_by_flags dev_get_by_name __dev_get_by_name dev_get_by_index __dev_get_by_index dev_ioctl dev_ethtool dev_load wireless_process_ioctl were modified to take a network namespace argument, and deal with it. vlan_ioctl_set and brioctl_set were modified so their hooks will receive a network namespace argument. So basically anthing in the core of the network stack that was affected to by the change of dev_base was modified to handle multiple network namespaces. The rest of the network stack was simply modified to explicitly use &init_net the initial network namespace. This can be fixed when those components of the network stack are modified to handle multiple network namespaces. For now the ifindex generator is left global. Fundametally ifindex numbers are per namespace, or else we will have corner case problems with migration when we get that far. At the same time there are assumptions in the network stack that the ifindex of a network device won't change. Making the ifindex number global seems a good compromise until the network stack can cope with ifindex changes when you change namespaces, and the like. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-18 02:56:21 +08:00
for_each_netdev(&init_net, dev)
dn_dev_down(dev);
rtnl_unlock();
}
void dn_dev_devices_on(void)
{
struct net_device *dev;
rtnl_lock();
[NET]: Make the device list and device lookups per namespace. This patch makes most of the generic device layer network namespace safe. This patch makes dev_base_head a network namespace variable, and then it picks up a few associated variables. The functions: dev_getbyhwaddr dev_getfirsthwbytype dev_get_by_flags dev_get_by_name __dev_get_by_name dev_get_by_index __dev_get_by_index dev_ioctl dev_ethtool dev_load wireless_process_ioctl were modified to take a network namespace argument, and deal with it. vlan_ioctl_set and brioctl_set were modified so their hooks will receive a network namespace argument. So basically anthing in the core of the network stack that was affected to by the change of dev_base was modified to handle multiple network namespaces. The rest of the network stack was simply modified to explicitly use &init_net the initial network namespace. This can be fixed when those components of the network stack are modified to handle multiple network namespaces. For now the ifindex generator is left global. Fundametally ifindex numbers are per namespace, or else we will have corner case problems with migration when we get that far. At the same time there are assumptions in the network stack that the ifindex of a network device won't change. Making the ifindex number global seems a good compromise until the network stack can cope with ifindex changes when you change namespaces, and the like. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-18 02:56:21 +08:00
for_each_netdev(&init_net, dev) {
if (dev->flags & IFF_UP)
dn_dev_up(dev);
}
rtnl_unlock();
}
int register_dnaddr_notifier(struct notifier_block *nb)
{
[PATCH] Notifier chain update: API changes The kernel's implementation of notifier chains is unsafe. There is no protection against entries being added to or removed from a chain while the chain is in use. The issues were discussed in this thread: http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2 We noticed that notifier chains in the kernel fall into two basic usage classes: "Blocking" chains are always called from a process context and the callout routines are allowed to sleep; "Atomic" chains can be called from an atomic context and the callout routines are not allowed to sleep. We decided to codify this distinction and make it part of the API. Therefore this set of patches introduces three new, parallel APIs: one for blocking notifiers, one for atomic notifiers, and one for "raw" notifiers (which is really just the old API under a new name). New kinds of data structures are used for the heads of the chains, and new routines are defined for registration, unregistration, and calling a chain. The three APIs are explained in include/linux/notifier.h and their implementation is in kernel/sys.c. With atomic and blocking chains, the implementation guarantees that the chain links will not be corrupted and that chain callers will not get messed up by entries being added or removed. For raw chains the implementation provides no guarantees at all; users of this API must provide their own protections. (The idea was that situations may come up where the assumptions of the atomic and blocking APIs are not appropriate, so it should be possible for users to handle these things in their own way.) There are some limitations, which should not be too hard to live with. For atomic/blocking chains, registration and unregistration must always be done in a process context since the chain is protected by a mutex/rwsem. Also, a callout routine for a non-raw chain must not try to register or unregister entries on its own chain. (This did happen in a couple of places and the code had to be changed to avoid it.) Since atomic chains may be called from within an NMI handler, they cannot use spinlocks for synchronization. Instead we use RCU. The overhead falls almost entirely in the unregister routine, which is okay since unregistration is much less frequent that calling a chain. Here is the list of chains that we adjusted and their classifications. None of them use the raw API, so for the moment it is only a placeholder. ATOMIC CHAINS ------------- arch/i386/kernel/traps.c: i386die_chain arch/ia64/kernel/traps.c: ia64die_chain arch/powerpc/kernel/traps.c: powerpc_die_chain arch/sparc64/kernel/traps.c: sparc64die_chain arch/x86_64/kernel/traps.c: die_chain drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list kernel/panic.c: panic_notifier_list kernel/profile.c: task_free_notifier net/bluetooth/hci_core.c: hci_notifier net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain net/ipv6/addrconf.c: inet6addr_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain net/netlink/af_netlink.c: netlink_chain BLOCKING CHAINS --------------- arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain arch/s390/kernel/process.c: idle_chain arch/x86_64/kernel/process.c idle_notifier drivers/base/memory.c: memory_chain drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list drivers/macintosh/adb.c: adb_client_list drivers/macintosh/via-pmu.c sleep_notifier_list drivers/macintosh/via-pmu68k.c sleep_notifier_list drivers/macintosh/windfarm_core.c wf_client_list drivers/usb/core/notify.c usb_notifier_list drivers/video/fbmem.c fb_notifier_list kernel/cpu.c cpu_chain kernel/module.c module_notify_list kernel/profile.c munmap_notifier kernel/profile.c task_exit_notifier kernel/sys.c reboot_notifier_list net/core/dev.c netdev_chain net/decnet/dn_dev.c: dnaddr_chain net/ipv4/devinet.c: inetaddr_chain It's possible that some of these classifications are wrong. If they are, please let us know or submit a patch to fix them. Note that any chain that gets called very frequently should be atomic, because the rwsem read-locking used for blocking chains is very likely to incur cache misses on SMP systems. (However, if the chain's callout routines may sleep then the chain cannot be atomic.) The patch set was written by Alan Stern and Chandra Seetharaman, incorporating material written by Keith Owens and suggestions from Paul McKenney and Andrew Morton. [jes@sgi.com: restructure the notifier chain initialization macros] Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com> Signed-off-by: Jes Sorensen <jes@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-27 17:16:30 +08:00
return blocking_notifier_chain_register(&dnaddr_chain, nb);
}
int unregister_dnaddr_notifier(struct notifier_block *nb)
{
[PATCH] Notifier chain update: API changes The kernel's implementation of notifier chains is unsafe. There is no protection against entries being added to or removed from a chain while the chain is in use. The issues were discussed in this thread: http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2 We noticed that notifier chains in the kernel fall into two basic usage classes: "Blocking" chains are always called from a process context and the callout routines are allowed to sleep; "Atomic" chains can be called from an atomic context and the callout routines are not allowed to sleep. We decided to codify this distinction and make it part of the API. Therefore this set of patches introduces three new, parallel APIs: one for blocking notifiers, one for atomic notifiers, and one for "raw" notifiers (which is really just the old API under a new name). New kinds of data structures are used for the heads of the chains, and new routines are defined for registration, unregistration, and calling a chain. The three APIs are explained in include/linux/notifier.h and their implementation is in kernel/sys.c. With atomic and blocking chains, the implementation guarantees that the chain links will not be corrupted and that chain callers will not get messed up by entries being added or removed. For raw chains the implementation provides no guarantees at all; users of this API must provide their own protections. (The idea was that situations may come up where the assumptions of the atomic and blocking APIs are not appropriate, so it should be possible for users to handle these things in their own way.) There are some limitations, which should not be too hard to live with. For atomic/blocking chains, registration and unregistration must always be done in a process context since the chain is protected by a mutex/rwsem. Also, a callout routine for a non-raw chain must not try to register or unregister entries on its own chain. (This did happen in a couple of places and the code had to be changed to avoid it.) Since atomic chains may be called from within an NMI handler, they cannot use spinlocks for synchronization. Instead we use RCU. The overhead falls almost entirely in the unregister routine, which is okay since unregistration is much less frequent that calling a chain. Here is the list of chains that we adjusted and their classifications. None of them use the raw API, so for the moment it is only a placeholder. ATOMIC CHAINS ------------- arch/i386/kernel/traps.c: i386die_chain arch/ia64/kernel/traps.c: ia64die_chain arch/powerpc/kernel/traps.c: powerpc_die_chain arch/sparc64/kernel/traps.c: sparc64die_chain arch/x86_64/kernel/traps.c: die_chain drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list kernel/panic.c: panic_notifier_list kernel/profile.c: task_free_notifier net/bluetooth/hci_core.c: hci_notifier net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain net/ipv6/addrconf.c: inet6addr_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain net/netlink/af_netlink.c: netlink_chain BLOCKING CHAINS --------------- arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain arch/s390/kernel/process.c: idle_chain arch/x86_64/kernel/process.c idle_notifier drivers/base/memory.c: memory_chain drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list drivers/macintosh/adb.c: adb_client_list drivers/macintosh/via-pmu.c sleep_notifier_list drivers/macintosh/via-pmu68k.c sleep_notifier_list drivers/macintosh/windfarm_core.c wf_client_list drivers/usb/core/notify.c usb_notifier_list drivers/video/fbmem.c fb_notifier_list kernel/cpu.c cpu_chain kernel/module.c module_notify_list kernel/profile.c munmap_notifier kernel/profile.c task_exit_notifier kernel/sys.c reboot_notifier_list net/core/dev.c netdev_chain net/decnet/dn_dev.c: dnaddr_chain net/ipv4/devinet.c: inetaddr_chain It's possible that some of these classifications are wrong. If they are, please let us know or submit a patch to fix them. Note that any chain that gets called very frequently should be atomic, because the rwsem read-locking used for blocking chains is very likely to incur cache misses on SMP systems. (However, if the chain's callout routines may sleep then the chain cannot be atomic.) The patch set was written by Alan Stern and Chandra Seetharaman, incorporating material written by Keith Owens and suggestions from Paul McKenney and Andrew Morton. [jes@sgi.com: restructure the notifier chain initialization macros] Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com> Signed-off-by: Jes Sorensen <jes@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-27 17:16:30 +08:00
return blocking_notifier_chain_unregister(&dnaddr_chain, nb);
}
#ifdef CONFIG_PROC_FS
static inline int is_dn_dev(struct net_device *dev)
{
return dev->dn_ptr != NULL;
}
static void *dn_dev_seq_start(struct seq_file *seq, loff_t *pos)
__acquires(RCU)
{
int i;
struct net_device *dev;
rcu_read_lock();
if (*pos == 0)
return SEQ_START_TOKEN;
i = 1;
for_each_netdev_rcu(&init_net, dev) {
if (!is_dn_dev(dev))
continue;
if (i++ == *pos)
return dev;
}
return NULL;
}
static void *dn_dev_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
struct net_device *dev;
++*pos;
dev = v;
if (v == SEQ_START_TOKEN)
[NET]: Make the device list and device lookups per namespace. This patch makes most of the generic device layer network namespace safe. This patch makes dev_base_head a network namespace variable, and then it picks up a few associated variables. The functions: dev_getbyhwaddr dev_getfirsthwbytype dev_get_by_flags dev_get_by_name __dev_get_by_name dev_get_by_index __dev_get_by_index dev_ioctl dev_ethtool dev_load wireless_process_ioctl were modified to take a network namespace argument, and deal with it. vlan_ioctl_set and brioctl_set were modified so their hooks will receive a network namespace argument. So basically anthing in the core of the network stack that was affected to by the change of dev_base was modified to handle multiple network namespaces. The rest of the network stack was simply modified to explicitly use &init_net the initial network namespace. This can be fixed when those components of the network stack are modified to handle multiple network namespaces. For now the ifindex generator is left global. Fundametally ifindex numbers are per namespace, or else we will have corner case problems with migration when we get that far. At the same time there are assumptions in the network stack that the ifindex of a network device won't change. Making the ifindex number global seems a good compromise until the network stack can cope with ifindex changes when you change namespaces, and the like. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-18 02:56:21 +08:00
dev = net_device_entry(&init_net.dev_base_head);
for_each_netdev_continue_rcu(&init_net, dev) {
if (!is_dn_dev(dev))
continue;
return dev;
}
return NULL;
}
static void dn_dev_seq_stop(struct seq_file *seq, void *v)
__releases(RCU)
{
rcu_read_unlock();
}
static char *dn_type2asc(char type)
{
switch (type) {
case DN_DEV_BCAST:
return "B";
case DN_DEV_UCAST:
return "U";
case DN_DEV_MPOINT:
return "M";
}
return "?";
}
static int dn_dev_seq_show(struct seq_file *seq, void *v)
{
if (v == SEQ_START_TOKEN)
seq_puts(seq, "Name Flags T1 Timer1 T3 Timer3 BlkSize Pri State DevType Router Peer\n");
else {
struct net_device *dev = v;
char peer_buf[DN_ASCBUF_LEN];
char router_buf[DN_ASCBUF_LEN];
struct dn_dev *dn_db = rcu_dereference(dev->dn_ptr);
seq_printf(seq, "%-8s %1s %04u %04u %04lu %04lu"
" %04hu %03d %02x %-10s %-7s %-7s\n",
dev->name ? dev->name : "???",
dn_type2asc(dn_db->parms.mode),
0, 0,
dn_db->t3, dn_db->parms.t3,
mtu2blksize(dev),
dn_db->parms.priority,
dn_db->parms.state, dn_db->parms.name,
dn_db->router ? dn_addr2asc(le16_to_cpu(*(__le16 *)dn_db->router->primary_key), router_buf) : "",
dn_db->peer ? dn_addr2asc(le16_to_cpu(*(__le16 *)dn_db->peer->primary_key), peer_buf) : "");
}
return 0;
}
static const struct seq_operations dn_dev_seq_ops = {
.start = dn_dev_seq_start,
.next = dn_dev_seq_next,
.stop = dn_dev_seq_stop,
.show = dn_dev_seq_show,
};
static int dn_dev_seq_open(struct inode *inode, struct file *file)
{
return seq_open(file, &dn_dev_seq_ops);
}
static const struct file_operations dn_dev_seq_fops = {
.owner = THIS_MODULE,
.open = dn_dev_seq_open,
.read = seq_read,
.llseek = seq_lseek,
.release = seq_release,
};
#endif /* CONFIG_PROC_FS */
[DECNET]: "addr" module param can't be __initdata sysfs keeps references to module parameters via /sys/module/*/parameters, so marking them as __initdata can't work. Steps to reproduce: modprobe decnet cat /sys/module/decnet/parameters/addr BUG: unable to handle kernel paging request at virtual address f88cd410 printing eip: c043dfd1 *pdpt = 0000000000004001 *pde = 0000000004408067 *pte = 0000000000000000 Oops: 0000 [#1] PREEMPT SMP Modules linked in: decnet sunrpc af_packet ipv6 binfmt_misc dm_mirror dm_multipath dm_mod sbs sbshc fan dock battery backlight ac power_supply parport loop rtc_cmos serio_raw rtc_core rtc_lib button amd_rng sr_mod cdrom shpchp pci_hotplug ehci_hcd ohci_hcd uhci_hcd usbcore Pid: 2099, comm: cat Not tainted (2.6.24-rc1-b1d08ac064268d0ae2281e98bf5e82627e0f0c56-bloat #6) EIP: 0060:[<c043dfd1>] EFLAGS: 00210286 CPU: 1 EIP is at param_get_int+0x6/0x20 EAX: c5c87000 EBX: 00000000 ECX: 000080d0 EDX: f88cd410 ESI: f8a108f8 EDI: c5c87000 EBP: 00000000 ESP: c5c97f00 DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 Process cat (pid: 2099, ti=c5c97000 task=c641ee10 task.ti=c5c97000) Stack: 00000000 f8a108f8 c5c87000 c043db6b f8a108f1 00000124 c043de1a c043db2f f88cd410 ffffffff c5c87000 f8a16bc8 f8a16bc8 c043dd69 c043dd54 c5dd5078 c043dbc8 c5cc7580 c06ee64c c5d679f8 c04c431f c641f480 c641f484 00001000 Call Trace: [<c043db6b>] param_array_get+0x3c/0x62 [<c043de1a>] param_array_set+0x0/0xdf [<c043db2f>] param_array_get+0x0/0x62 [<c043dd69>] param_attr_show+0x15/0x2d [<c043dd54>] param_attr_show+0x0/0x2d [<c043dbc8>] module_attr_show+0x1a/0x1e [<c04c431f>] sysfs_read_file+0x7c/0xd9 [<c04c42a3>] sysfs_read_file+0x0/0xd9 [<c048d4b2>] vfs_read+0x88/0x134 [<c042090b>] do_page_fault+0x0/0x7d5 [<c048d920>] sys_read+0x41/0x67 [<c04080fa>] sysenter_past_esp+0x6b/0xc1 ======================= Code: 00 83 c4 0c c3 83 ec 0c 8b 52 10 8b 12 c7 44 24 04 27 dd 6c c0 89 04 24 89 54 24 08 e8 ea 01 0c 00 83 c4 0c c3 83 ec 0c 8b 52 10 <8b> 12 c7 44 24 04 58 8c 6a c0 89 04 24 89 54 24 08 e8 ca 01 0c EIP: [<c043dfd1>] param_get_int+0x6/0x20 SS:ESP 0068:c5c97f00 Signed-off-by: Alexey Dobriyan <adobriyan@sw.ru> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-06 13:30:11 +08:00
static int addr[2];
module_param_array(addr, int, NULL, 0444);
MODULE_PARM_DESC(addr, "The DECnet address of this machine: area,node");
void __init dn_dev_init(void)
{
if (addr[0] > 63 || addr[0] < 0) {
printk(KERN_ERR "DECnet: Area must be between 0 and 63");
return;
}
if (addr[1] > 1023 || addr[1] < 0) {
printk(KERN_ERR "DECnet: Node must be between 0 and 1023");
return;
}
decnet_address = cpu_to_le16((addr[0] << 10) | addr[1]);
dn_dev_devices_on();
rtnl_register(PF_DECnet, RTM_NEWADDR, dn_nl_newaddr, NULL, NULL);
rtnl_register(PF_DECnet, RTM_DELADDR, dn_nl_deladdr, NULL, NULL);
rtnl_register(PF_DECnet, RTM_GETADDR, NULL, dn_nl_dump_ifaddr, NULL);
proc_create("decnet_dev", S_IRUGO, init_net.proc_net, &dn_dev_seq_fops);
#ifdef CONFIG_SYSCTL
{
int i;
for(i = 0; i < DN_DEV_LIST_SIZE; i++)
dn_dev_sysctl_register(NULL, &dn_dev_list[i]);
}
#endif /* CONFIG_SYSCTL */
}
void __exit dn_dev_cleanup(void)
{
#ifdef CONFIG_SYSCTL
{
int i;
for(i = 0; i < DN_DEV_LIST_SIZE; i++)
dn_dev_sysctl_unregister(&dn_dev_list[i]);
}
#endif /* CONFIG_SYSCTL */
remove_proc_entry("decnet_dev", init_net.proc_net);
dn_dev_devices_off();
}