2005-04-16 18:20:36 -04:00
|
|
|
/*
|
|
|
|
* net/core/dst.c Protocol independent destination cache.
|
|
|
|
*
|
|
|
|
* Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/bitops.h>
|
|
|
|
#include <linux/errno.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/kernel.h>
|
2007-09-12 08:29:01 -04:00
|
|
|
#include <linux/workqueue.h>
|
2005-04-16 18:20:36 -04:00
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/module.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 04:04:11 -04:00
|
|
|
#include <linux/slab.h>
|
2005-04-16 18:20:36 -04:00
|
|
|
#include <linux/netdevice.h>
|
|
|
|
#include <linux/skbuff.h>
|
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/types.h>
|
2007-09-12 07:02:17 -04:00
|
|
|
#include <net/net_namespace.h>
|
2010-02-08 18:00:39 -05:00
|
|
|
#include <linux/sched.h>
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
#include <net/dst.h>
|
|
|
|
|
2007-09-12 08:29:01 -04:00
|
|
|
/*
|
|
|
|
* Theory of operations:
|
|
|
|
* 1) We use a list, protected by a spinlock, to add
|
|
|
|
* new entries from both BH and non-BH context.
|
|
|
|
* 2) In order to keep spinlock held for a small delay,
|
|
|
|
* we use a second list where are stored long lived
|
|
|
|
* entries, that are handled by the garbage collect thread
|
|
|
|
* fired by a workqueue.
|
|
|
|
* 3) This list is guarded by a mutex,
|
|
|
|
* so that the gc_task and dst_dev_event() can be synchronized.
|
2005-04-16 18:20:36 -04:00
|
|
|
*/
|
2007-02-09 09:24:36 -05:00
|
|
|
#if RT_CACHE_DEBUG >= 2
|
2005-04-16 18:20:36 -04:00
|
|
|
static atomic_t dst_total = ATOMIC_INIT(0);
|
|
|
|
#endif
|
|
|
|
|
2007-09-12 08:29:01 -04:00
|
|
|
/*
|
|
|
|
* We want to keep lock & list close together
|
|
|
|
* to dirty as few cache lines as possible in __dst_free().
|
|
|
|
* As this is not a very strong hint, we dont force an alignment on SMP.
|
|
|
|
*/
|
|
|
|
static struct {
|
|
|
|
spinlock_t lock;
|
2010-03-29 06:41:36 -04:00
|
|
|
struct dst_entry *list;
|
2007-09-12 08:29:01 -04:00
|
|
|
unsigned long timer_inc;
|
|
|
|
unsigned long timer_expires;
|
|
|
|
} dst_garbage = {
|
|
|
|
.lock = __SPIN_LOCK_UNLOCKED(dst_garbage.lock),
|
|
|
|
.timer_inc = DST_GC_MAX,
|
|
|
|
};
|
|
|
|
static void dst_gc_task(struct work_struct *work);
|
2010-03-29 06:41:36 -04:00
|
|
|
static void ___dst_free(struct dst_entry *dst);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2007-09-12 08:29:01 -04:00
|
|
|
static DECLARE_DELAYED_WORK(dst_gc_work, dst_gc_task);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2007-09-12 08:29:01 -04:00
|
|
|
static DEFINE_MUTEX(dst_gc_mutex);
|
|
|
|
/*
|
|
|
|
* long lived entries are maintained in this list, guarded by dst_gc_mutex
|
|
|
|
*/
|
|
|
|
static struct dst_entry *dst_busy_list;
|
|
|
|
|
|
|
|
static void dst_gc_task(struct work_struct *work)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
int delayed = 0;
|
2007-09-12 08:29:01 -04:00
|
|
|
int work_performed = 0;
|
|
|
|
unsigned long expires = ~0L;
|
|
|
|
struct dst_entry *dst, *next, head;
|
|
|
|
struct dst_entry *last = &head;
|
|
|
|
#if RT_CACHE_DEBUG >= 2
|
|
|
|
ktime_t time_start = ktime_get();
|
|
|
|
struct timespec elapsed;
|
|
|
|
#endif
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2007-09-12 08:29:01 -04:00
|
|
|
mutex_lock(&dst_gc_mutex);
|
|
|
|
next = dst_busy_list;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2007-09-12 08:29:01 -04:00
|
|
|
loop:
|
|
|
|
while ((dst = next) != NULL) {
|
|
|
|
next = dst->next;
|
|
|
|
prefetch(&next->next);
|
2010-02-08 18:00:39 -05:00
|
|
|
cond_resched();
|
2007-09-12 08:29:01 -04:00
|
|
|
if (likely(atomic_read(&dst->__refcnt))) {
|
|
|
|
last->next = dst;
|
|
|
|
last = dst;
|
2005-04-16 18:20:36 -04:00
|
|
|
delayed++;
|
|
|
|
continue;
|
|
|
|
}
|
2007-09-12 08:29:01 -04:00
|
|
|
work_performed++;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
dst = dst_destroy(dst);
|
|
|
|
if (dst) {
|
|
|
|
/* NOHASH and still referenced. Unless it is already
|
|
|
|
* on gc list, invalidate it and add to gc list.
|
|
|
|
*
|
|
|
|
* Note: this is temporary. Actually, NOHASH dst's
|
|
|
|
* must be obsoleted when parent is obsoleted.
|
|
|
|
* But we do not have state "obsoleted, but
|
|
|
|
* referenced by parent", so it is right.
|
|
|
|
*/
|
|
|
|
if (dst->obsolete > 1)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
___dst_free(dst);
|
2007-09-12 08:29:01 -04:00
|
|
|
dst->next = next;
|
|
|
|
next = dst;
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
|
|
|
}
|
2007-09-12 08:29:01 -04:00
|
|
|
|
|
|
|
spin_lock_bh(&dst_garbage.lock);
|
|
|
|
next = dst_garbage.list;
|
|
|
|
if (next) {
|
|
|
|
dst_garbage.list = NULL;
|
|
|
|
spin_unlock_bh(&dst_garbage.lock);
|
|
|
|
goto loop;
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
2007-09-12 08:29:01 -04:00
|
|
|
last->next = NULL;
|
|
|
|
dst_busy_list = head.next;
|
|
|
|
if (!dst_busy_list)
|
|
|
|
dst_garbage.timer_inc = DST_GC_MAX;
|
|
|
|
else {
|
|
|
|
/*
|
|
|
|
* if we freed less than 1/10 of delayed entries,
|
|
|
|
* we can sleep longer.
|
|
|
|
*/
|
|
|
|
if (work_performed <= delayed/10) {
|
|
|
|
dst_garbage.timer_expires += dst_garbage.timer_inc;
|
|
|
|
if (dst_garbage.timer_expires > DST_GC_MAX)
|
|
|
|
dst_garbage.timer_expires = DST_GC_MAX;
|
|
|
|
dst_garbage.timer_inc += DST_GC_INC;
|
|
|
|
} else {
|
|
|
|
dst_garbage.timer_inc = DST_GC_INC;
|
|
|
|
dst_garbage.timer_expires = DST_GC_MIN;
|
|
|
|
}
|
|
|
|
expires = dst_garbage.timer_expires;
|
|
|
|
/*
|
2010-03-29 06:41:36 -04:00
|
|
|
* if the next desired timer is more than 4 seconds in the
|
|
|
|
* future then round the timer to whole seconds
|
2007-09-12 08:29:01 -04:00
|
|
|
*/
|
|
|
|
if (expires > 4*HZ)
|
|
|
|
expires = round_jiffies_relative(expires);
|
|
|
|
schedule_delayed_work(&dst_gc_work, expires);
|
2005-07-30 20:47:25 -04:00
|
|
|
}
|
2007-09-12 08:29:01 -04:00
|
|
|
|
|
|
|
spin_unlock_bh(&dst_garbage.lock);
|
|
|
|
mutex_unlock(&dst_gc_mutex);
|
2005-04-16 18:20:36 -04:00
|
|
|
#if RT_CACHE_DEBUG >= 2
|
2007-09-12 08:29:01 -04:00
|
|
|
elapsed = ktime_to_timespec(ktime_sub(ktime_get(), time_start));
|
|
|
|
printk(KERN_DEBUG "dst_total: %d delayed: %d work_perf: %d"
|
|
|
|
" expires: %lu elapsed: %lu us\n",
|
|
|
|
atomic_read(&dst_total), delayed, work_performed,
|
|
|
|
expires,
|
2010-03-29 06:41:36 -04:00
|
|
|
elapsed.tv_sec * USEC_PER_SEC +
|
|
|
|
elapsed.tv_nsec / NSEC_PER_USEC);
|
2005-04-16 18:20:36 -04:00
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2007-11-14 00:34:06 -05:00
|
|
|
int dst_discard(struct sk_buff *skb)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
kfree_skb(skb);
|
|
|
|
return 0;
|
|
|
|
}
|
2007-11-14 00:34:06 -05:00
|
|
|
EXPORT_SYMBOL(dst_discard);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2010-03-29 06:41:36 -04:00
|
|
|
void *dst_alloc(struct dst_ops *ops)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
2010-03-29 06:41:36 -04:00
|
|
|
struct dst_entry *dst;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2010-10-08 02:37:34 -04:00
|
|
|
if (ops->gc && dst_entries_get_fast(ops) > ops->gc_thresh) {
|
2008-01-18 06:56:57 -05:00
|
|
|
if (ops->gc(ops))
|
2005-04-16 18:20:36 -04:00
|
|
|
return NULL;
|
|
|
|
}
|
2007-02-10 04:45:03 -05:00
|
|
|
dst = kmem_cache_zalloc(ops->kmem_cachep, GFP_ATOMIC);
|
2005-04-16 18:20:36 -04:00
|
|
|
if (!dst)
|
|
|
|
return NULL;
|
|
|
|
atomic_set(&dst->__refcnt, 0);
|
|
|
|
dst->ops = ops;
|
|
|
|
dst->lastuse = jiffies;
|
|
|
|
dst->path = dst;
|
2007-06-05 03:06:57 -04:00
|
|
|
dst->input = dst->output = dst_discard;
|
2007-02-09 09:24:36 -05:00
|
|
|
#if RT_CACHE_DEBUG >= 2
|
2005-04-16 18:20:36 -04:00
|
|
|
atomic_inc(&dst_total);
|
|
|
|
#endif
|
2010-10-08 02:37:34 -04:00
|
|
|
dst_entries_add(ops, 1);
|
2005-04-16 18:20:36 -04:00
|
|
|
return dst;
|
|
|
|
}
|
2010-03-29 06:41:36 -04:00
|
|
|
EXPORT_SYMBOL(dst_alloc);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2010-03-29 06:41:36 -04:00
|
|
|
static void ___dst_free(struct dst_entry *dst)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
/* The first case (dev==NULL) is required, when
|
|
|
|
protocol module is unloaded.
|
|
|
|
*/
|
2010-03-29 06:41:36 -04:00
|
|
|
if (dst->dev == NULL || !(dst->dev->flags&IFF_UP))
|
2007-06-05 03:06:57 -04:00
|
|
|
dst->input = dst->output = dst_discard;
|
2005-04-16 18:20:36 -04:00
|
|
|
dst->obsolete = 2;
|
|
|
|
}
|
|
|
|
|
2010-03-29 06:41:36 -04:00
|
|
|
void __dst_free(struct dst_entry *dst)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
2007-09-12 08:29:01 -04:00
|
|
|
spin_lock_bh(&dst_garbage.lock);
|
2005-04-16 18:20:36 -04:00
|
|
|
___dst_free(dst);
|
2007-09-12 08:29:01 -04:00
|
|
|
dst->next = dst_garbage.list;
|
|
|
|
dst_garbage.list = dst;
|
|
|
|
if (dst_garbage.timer_inc > DST_GC_INC) {
|
|
|
|
dst_garbage.timer_inc = DST_GC_INC;
|
|
|
|
dst_garbage.timer_expires = DST_GC_MIN;
|
net: fix scheduling of dst_gc_task by __dst_free
The dst garbage collector dst_gc_task() may not be scheduled as we
expect it to be in __dst_free().
Indeed, when the dst_gc_timer was replaced by the delayed_work
dst_gc_work, the mod_timer() call used to schedule the garbage
collector at an earlier date was replaced by a schedule_delayed_work()
(see commit 86bba269d08f0c545ae76c90b56727f65d62d57f).
But, the behaviour of mod_timer() and schedule_delayed_work() is
different in the way they handle the delay.
mod_timer() stops the timer and re-arm it with the new given delay,
whereas schedule_delayed_work() only check if the work is already
queued in the workqueue (and queue it (with delay) if it is not)
BUT it does NOT take into account the new delay (even if the new delay
is earlier in time).
schedule_delayed_work() returns 0 if it didn't queue the work,
but we don't check the return code in __dst_free().
If I understand the code in __dst_free() correctly, we want dst_gc_task
to be queued after DST_GC_INC jiffies if we pass the test (and not in
some undetermined time in the future), so I think we should add a call
to cancel_delayed_work() before schedule_delayed_work(). Patch below.
Or we should at least test the return code of schedule_delayed_work(),
and reset the values of dst_garbage.timer_inc and dst_garbage.timer_expires
back to their former values if schedule_delayed_work() failed.
Otherwise the subsequent calls to __dst_free will test the wrong values
and assume wrong thing about when the garbage collector is supposed to
be scheduled.
dst_gc_task() also calls schedule_delayed_work() without checking
its return code (or calling cancel_scheduled_work() first), but it
should fine there: dst_gc_task is the routine of the delayed_work, so
no dst_gc_work should be pending in the queue when it's running.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-09-12 19:16:37 -04:00
|
|
|
cancel_delayed_work(&dst_gc_work);
|
2007-09-12 08:29:01 -04:00
|
|
|
schedule_delayed_work(&dst_gc_work, dst_garbage.timer_expires);
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
2007-09-12 08:29:01 -04:00
|
|
|
spin_unlock_bh(&dst_garbage.lock);
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
2010-07-19 19:51:38 -04:00
|
|
|
EXPORT_SYMBOL(__dst_free);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
struct dst_entry *dst_destroy(struct dst_entry * dst)
|
|
|
|
{
|
|
|
|
struct dst_entry *child;
|
|
|
|
struct neighbour *neigh;
|
|
|
|
struct hh_cache *hh;
|
|
|
|
|
|
|
|
smp_rmb();
|
|
|
|
|
|
|
|
again:
|
|
|
|
neigh = dst->neighbour;
|
|
|
|
hh = dst->hh;
|
|
|
|
child = dst->child;
|
|
|
|
|
|
|
|
dst->hh = NULL;
|
2010-10-11 12:16:57 -04:00
|
|
|
if (hh)
|
|
|
|
hh_cache_put(hh);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
if (neigh) {
|
|
|
|
dst->neighbour = NULL;
|
|
|
|
neigh_release(neigh);
|
|
|
|
}
|
|
|
|
|
2010-10-08 02:37:34 -04:00
|
|
|
dst_entries_add(dst->ops, -1);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
if (dst->ops->destroy)
|
|
|
|
dst->ops->destroy(dst);
|
|
|
|
if (dst->dev)
|
|
|
|
dev_put(dst->dev);
|
2007-02-09 09:24:36 -05:00
|
|
|
#if RT_CACHE_DEBUG >= 2
|
2005-04-16 18:20:36 -04:00
|
|
|
atomic_dec(&dst_total);
|
|
|
|
#endif
|
|
|
|
kmem_cache_free(dst->ops->kmem_cachep, dst);
|
|
|
|
|
|
|
|
dst = child;
|
|
|
|
if (dst) {
|
2005-04-16 18:24:10 -04:00
|
|
|
int nohash = dst->flags & DST_NOHASH;
|
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
if (atomic_dec_and_test(&dst->__refcnt)) {
|
|
|
|
/* We were real parent of this dst, so kill child. */
|
2005-04-16 18:24:10 -04:00
|
|
|
if (nohash)
|
2005-04-16 18:20:36 -04:00
|
|
|
goto again;
|
|
|
|
} else {
|
|
|
|
/* Child is still referenced, return it for freeing. */
|
2005-04-16 18:24:10 -04:00
|
|
|
if (nohash)
|
2005-04-16 18:20:36 -04:00
|
|
|
return dst;
|
|
|
|
/* Child is still in his hash table */
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return NULL;
|
|
|
|
}
|
2010-03-29 06:41:36 -04:00
|
|
|
EXPORT_SYMBOL(dst_destroy);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2008-03-27 20:53:31 -04:00
|
|
|
void dst_release(struct dst_entry *dst)
|
|
|
|
{
|
|
|
|
if (dst) {
|
2010-03-29 06:41:36 -04:00
|
|
|
int newrefcnt;
|
2008-11-14 03:53:54 -05:00
|
|
|
|
2010-03-29 06:41:36 -04:00
|
|
|
newrefcnt = atomic_dec_return(&dst->__refcnt);
|
|
|
|
WARN_ON(newrefcnt < 0);
|
2010-10-15 01:44:11 -04:00
|
|
|
if (unlikely(dst->flags & DST_NOCACHE) && !newrefcnt) {
|
|
|
|
dst = dst_destroy(dst);
|
|
|
|
if (dst)
|
|
|
|
__dst_free(dst);
|
|
|
|
}
|
2008-03-27 20:53:31 -04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(dst_release);
|
|
|
|
|
2010-10-15 01:44:11 -04:00
|
|
|
/**
|
|
|
|
* skb_dst_set_noref - sets skb dst, without a reference
|
|
|
|
* @skb: buffer
|
|
|
|
* @dst: dst entry
|
|
|
|
*
|
|
|
|
* Sets skb dst, assuming a reference was not taken on dst
|
|
|
|
* skb_dst_drop() should not dst_release() this dst
|
|
|
|
*/
|
|
|
|
void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst)
|
|
|
|
{
|
|
|
|
WARN_ON(!rcu_read_lock_held() && !rcu_read_lock_bh_held());
|
|
|
|
/* If dst not in cache, we must take a reference, because
|
|
|
|
* dst_release() will destroy dst as soon as its refcount becomes zero
|
|
|
|
*/
|
|
|
|
if (unlikely(dst->flags & DST_NOCACHE)) {
|
|
|
|
dst_hold(dst);
|
|
|
|
skb_dst_set(skb, dst);
|
|
|
|
} else {
|
|
|
|
skb->_skb_refdst = (unsigned long)dst | SKB_DST_NOREF;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(skb_dst_set_noref);
|
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
/* Dirty hack. We did it in 2.2 (in __dst_free),
|
|
|
|
* we have _very_ good reasons not to repeat
|
|
|
|
* this mistake in 2.3, but we have no choice
|
|
|
|
* now. _It_ _is_ _explicit_ _deliberate_
|
|
|
|
* _race_ _condition_.
|
|
|
|
*
|
|
|
|
* Commented and originally written by Alexey.
|
|
|
|
*/
|
2010-04-12 03:38:05 -04:00
|
|
|
static void dst_ifdown(struct dst_entry *dst, struct net_device *dev,
|
|
|
|
int unregister)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
if (dst->ops->ifdown)
|
|
|
|
dst->ops->ifdown(dst, dev, unregister);
|
|
|
|
|
|
|
|
if (dev != dst->dev)
|
|
|
|
return;
|
|
|
|
|
|
|
|
if (!unregister) {
|
2007-06-05 03:06:57 -04:00
|
|
|
dst->input = dst->output = dst_discard;
|
2005-04-16 18:20:36 -04:00
|
|
|
} else {
|
2008-03-25 08:47:49 -04:00
|
|
|
dst->dev = dev_net(dst->dev)->loopback_dev;
|
2007-09-25 22:16:28 -04:00
|
|
|
dev_hold(dst->dev);
|
2005-04-16 18:20:36 -04:00
|
|
|
dev_put(dev);
|
|
|
|
if (dst->neighbour && dst->neighbour->dev == dev) {
|
2007-12-07 03:38:10 -05:00
|
|
|
dst->neighbour->dev = dst->dev;
|
2007-12-11 05:00:30 -05:00
|
|
|
dev_hold(dst->dev);
|
2005-04-16 18:20:36 -04:00
|
|
|
dev_put(dev);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-03-29 06:41:36 -04:00
|
|
|
static int dst_dev_event(struct notifier_block *this, unsigned long event,
|
|
|
|
void *ptr)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
struct net_device *dev = ptr;
|
2007-09-12 08:29:01 -04:00
|
|
|
struct dst_entry *dst, *last = NULL;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
switch (event) {
|
|
|
|
case NETDEV_UNREGISTER:
|
|
|
|
case NETDEV_DOWN:
|
2007-09-12 08:29:01 -04:00
|
|
|
mutex_lock(&dst_gc_mutex);
|
|
|
|
for (dst = dst_busy_list; dst; dst = dst->next) {
|
|
|
|
last = dst;
|
|
|
|
dst_ifdown(dst, dev, event != NETDEV_DOWN);
|
|
|
|
}
|
|
|
|
|
|
|
|
spin_lock_bh(&dst_garbage.lock);
|
|
|
|
dst = dst_garbage.list;
|
|
|
|
dst_garbage.list = NULL;
|
|
|
|
spin_unlock_bh(&dst_garbage.lock);
|
|
|
|
|
|
|
|
if (last)
|
|
|
|
last->next = dst;
|
|
|
|
else
|
|
|
|
dst_busy_list = dst;
|
2010-03-29 06:41:36 -04:00
|
|
|
for (; dst; dst = dst->next)
|
2005-04-16 18:20:36 -04:00
|
|
|
dst_ifdown(dst, dev, event != NETDEV_DOWN);
|
2007-09-12 08:29:01 -04:00
|
|
|
mutex_unlock(&dst_gc_mutex);
|
2005-04-16 18:20:36 -04:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
return NOTIFY_DONE;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct notifier_block dst_dev_notifier = {
|
|
|
|
.notifier_call = dst_dev_event,
|
2010-11-09 14:46:33 -05:00
|
|
|
.priority = -10, /* must be called after other network notifiers */
|
2005-04-16 18:20:36 -04:00
|
|
|
};
|
|
|
|
|
|
|
|
void __init dst_init(void)
|
|
|
|
{
|
|
|
|
register_netdevice_notifier(&dst_dev_notifier);
|
|
|
|
}
|