Skip to contentSkip to conversation

netfilter: don't xt_jumpstack_alloc twice in xt_register_table

PATCH Applied Xiaotian Feng 4 participants 11 messages May 31, 2010

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 netfilter: don't xt_jumpstack_alloc twice in xt_register_table

In xt_register_table, xt_jumpstack_alloc is called first, later xt_replace_table is used. But in xt_replace_table, xt_jumpstack_alloc will be used again. Then the memory allocated by previous xt_jumpstack_alloc will be leaked. We can simply remove the previous xt_jumpstack_alloc because there aren't any users of newinfo between xt_jumpstack_alloc and xt_replace_table.

Cc Patrick McHardy <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Jan Engelhardt <[email protected]>
Cc Andrew Morton <[email protected]>
Cc Rusty Russell <[email protected]>
Cc Alexey Dobriyan <[email protected]>
---
 net/netfilter/x_tables.c |    4 ----
 1 files changed, 0 insertions(+), 4 deletions(-)
net/netfilter/x_tables.c +0 -4
index 445de70..47b1e79 100644
@@ -844,10 +844,6 @@ struct xt_table *xt_register_table(struct net *net,
 	struct xt_table_info *private;
 	struct xt_table *t, *table;
 
-	ret = xt_jumpstack_alloc(newinfo);
-	if (ret < 0)
-		return ERR_PTR(ret);
-
 	/* Don't add one object to multiple lists. */
 	table = kmemdup(input_table, sizeof(struct xt_table), GFP_KERNEL);
 	if (!table) {
Signature · 1 line
1.7.0.1
#1 lore raw
Quoted 6 lines · Xiaotian Feng

In xt_register_table, xt_jumpstack_alloc is called first, later xt_replace_table is used. But in xt_replace_table, xt_jumpstack_alloc will be used again. Then the memory allocated by previous xt_jumpstack_alloc will be leaked. We can simply remove the previous xt_jumpstack_alloc because there aren't any users of newinfo between xt_jumpstack_alloc and xt_replace_table.

Indeed that seems to be so.

Quoted 15 lines
net/netfilter/x_tables.c +0 -4
index 445de70..47b1e79 100644
@@ -844,10 +844,6 @@ struct xt_table *xt_register_table(struct net *net,
 	struct xt_table_info *private;
 	struct xt_table *t, *table;
 
-	ret = xt_jumpstack_alloc(newinfo);
-	if (ret < 0)
-		return ERR_PTR(ret);
-
 	/* Don't add one object to multiple lists. */
 	table = kmemdup(input_table, sizeof(struct xt_table), GFP_KERNEL);
 	if (!table) {
#2 lore raw

PATCH 1/1 netfilter: xtables: stackptr should be percpu

Quoted 10 lines · Jan Engelhardt
Quoted 6 lines · Xiaotian Feng

In xt_register_table, xt_jumpstack_alloc is called first, later xt_replace_table is used. But in xt_replace_table, xt_jumpstack_alloc will be used again. Then the memory allocated by previous xt_jumpstack_alloc will be leaked. We can simply remove the previous xt_jumpstack_alloc because there aren't any users of newinfo between xt_jumpstack_alloc and xt_replace_table.

Indeed that seems to be so.

An official "Acked-by: ..." would be fine Jan :)

BTW I noticed a _big_ slowdown of iptables lately, and located the reason.

All cpus share a single cache line for their 'stackptr' storage, introduced in commit f3c5c1bfd4

This is a stable candidate (2.6.34)

Note : We also should use alloc_percpu() for jumpstack but this is not a critical thing and can be a net-next patch.

[PATCH] netfilter: xtables: stackptr should be percpu

commit f3c5c1bfd4 (netfilter: xtables: make ip_tables reentrant) introduced a performance regression, because stackptr array is shared by all cpus, adding cache line ping pongs. (16 cpus share a 64 bytes cache line)

Fix this using alloc_percpu()

---
 include/linux/netfilter/x_tables.h |    2 +-
 net/ipv4/netfilter/ip_tables.c     |    2 +-
 net/ipv6/netfilter/ip6_tables.c    |    2 +-
 net/netfilter/x_tables.c           |   13 +++----------
 4 files changed, 6 insertions(+), 13 deletions(-)
include/linux/netfilter/x_tables.h +1 -1
index c00cc0c..24e5d01 100644
@@ -397,7 +397,7 @@ struct xt_table_info {
 	 * @stacksize jumps (number of user chains) can possibly be made.
 	 */
 	unsigned int stacksize;
-	unsigned int *stackptr;
+	unsigned int __percpu *stackptr;
 	void ***jumpstack;
 	/* ipt_entry tables: one per CPU */
 	/* Note : this field MUST be the last one, see XT_TABLE_INFO_SZ */
net/ipv4/netfilter/ip_tables.c +1 -1
index 63958f3..4b6c5ca 100644
@@ -336,7 +336,7 @@ ipt_do_table(struct sk_buff *skb,
 	cpu        = smp_processor_id();
 	table_base = private->entries[cpu];
 	jumpstack  = (struct ipt_entry **)private->jumpstack[cpu];
-	stackptr   = &private->stackptr[cpu];
+	stackptr   = per_cpu_ptr(private->stackptr, cpu);
 	origptr    = *stackptr;
 
 	e = get_entry(table_base, private->hook_entry[hook]);
net/ipv6/netfilter/ip6_tables.c +1 -1
index 6f517bd..9d2d68f 100644
@@ -363,7 +363,7 @@ ip6t_do_table(struct sk_buff *skb,
 	cpu        = smp_processor_id();
 	table_base = private->entries[cpu];
 	jumpstack  = (struct ip6t_entry **)private->jumpstack[cpu];
-	stackptr   = &private->stackptr[cpu];
+	stackptr   = per_cpu_ptr(private->stackptr, cpu);
 	origptr    = *stackptr;
 
 	e = get_entry(table_base, private->hook_entry[hook]);
net/netfilter/x_tables.c +3 -10
index 445de70..7e8a93d 100644
@@ -699,10 +699,8 @@ void xt_free_table_info(struct xt_table_info *info)
 		vfree(info->jumpstack);
 	else
 		kfree(info->jumpstack);
-	if (sizeof(unsigned int) * nr_cpu_ids > PAGE_SIZE)
-		vfree(info->stackptr);
-	else
-		kfree(info->stackptr);
+
+	free_percpu(info->stackptr);
 
 	kfree(info);
 }
@@ -753,14 +751,9 @@ static int xt_jumpstack_alloc(struct xt_table_info *i)
 	unsigned int size;
 	int cpu;
 
-	size = sizeof(unsigned int) * nr_cpu_ids;
-	if (size > PAGE_SIZE)
-		i->stackptr = vmalloc(size);
-	else
-		i->stackptr = kmalloc(size, GFP_KERNEL);
+	i->stackptr = alloc_percpu(unsigned int);
 	if (i->stackptr == NULL)
 		return -ENOMEM;
-	memset(i->stackptr, 0, size);
 
 	size = sizeof(void **) * nr_cpu_ids;
 	if (size > PAGE_SIZE)
Acked-by on v1 1/1
#3 lore raw

Re: [PATCH] netfilter: xtables: stackptr should be percpu

Quoted 5 lines · Eric Dumazet

All cpus share a single cache line for their 'stackptr' storage, introduced in commit f3c5c1bfd4

This is a stable candidate (2.6.34)

Stackptr was first introduced for 2.6.35-rcX.

Quoted 7 lines
net/netfilter/x_tables.c +1 -0
+	i->stackptr = alloc_percpu(unsigned int);
	if (i->stackptr == NULL)
		return -ENOMEM;
net/netfilter/x_tables.c +0 -1
-	memset(i->stackptr, 0, size);
	size = sizeof(void **) * nr_cpu_ids;
	if (size > PAGE_SIZE)

Are alloc_percpu areas cleared?

#4 lore raw

Re: [PATCH] netfilter: xtables: stackptr should be percpu

Quoted 9 lines · Jan Engelhardt
Quoted 5 lines · Eric Dumazet

All cpus share a single cache line for their 'stackptr' storage, introduced in commit f3c5c1bfd4

This is a stable candidate (2.6.34)

Stackptr was first introduced for 2.6.35-rcX.

Indeed, I was fooled by 'git describe'

Quoted 10 lines
Quoted 7 lines
net/netfilter/x_tables.c +1 -0
+	i->stackptr = alloc_percpu(unsigned int);
	if (i->stackptr == NULL)
		return -ENOMEM;
net/netfilter/x_tables.c +0 -1
-	memset(i->stackptr, 0, size);
	size = sizeof(void **) * nr_cpu_ids;
	if (size > PAGE_SIZE)

Are alloc_percpu areas cleared?

Yes, allocated chunks are cleared.

--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#5 lore raw

Re: [PATCH] netfilter: xtables: stackptr should be percpu

Quoted 11 lines · Eric Dumazet
Quoted 8 lines · Jan Engelhardt
Quoted 5 lines · Eric Dumazet

All cpus share a single cache line for their 'stackptr' storage, introduced in commit f3c5c1bfd4

This is a stable candidate (2.6.34)

Stackptr was first introduced for 2.6.35-rcX.

Indeed, I was fooled by 'git describe'

Keep your friends close, and your enemies closer ;-)

	git describe --contains f3c5c1bfd

--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#6 lore raw

Re: [PATCH] netfilter: xtables: stackptr should be percpu

Quoted 3 lines · Jan Engelhardt

Keep your friends close, and your enemies closer ;-)

	git describe --contains f3c5c1bfd

Yes, --contains should be the default, and --predates the option :)

This is a bit OT anyway :)

--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#7 lore raw

Re: [PATCH] netfilter: xtables: stackptr should be percpu

In reply to Jan Engelhardt
Quoted 13 lines · Jan Engelhardt
Quoted 4 lines · Eric Dumazet

All cpus share a single cache line for their 'stackptr' storage, introduced in commit f3c5c1bfd4

This is a stable candidate (2.6.34)

Stackptr was first introduced for 2.6.35-rcX.

Quoted 5 lines
net/netfilter/x_tables.c +1 -0
+	i->stackptr = alloc_percpu(unsigned int);
	if (i->stackptr == NULL)
		return -ENOMEM;
net/netfilter/x_tables.c +0 -1
-	memset(i->stackptr, 0, size);
Quoted 6 lines
Quoted 2 lines
	size = sizeof(void **) * nr_cpu_ids;
	if (size > PAGE_SIZE)

Are alloc_percpu areas cleared?

Applied, thanks.

#8 lore raw
In reply to Jan Engelhardt
Quoted 10 lines · Jan Engelhardt
Quoted 6 lines · Xiaotian Feng

In xt_register_table, xt_jumpstack_alloc is called first, later xt_replace_table is used. But in xt_replace_table, xt_jumpstack_alloc will be used again. Then the memory allocated by previous xt_jumpstack_alloc will be leaked. We can simply remove the previous xt_jumpstack_alloc because there aren't any users of newinfo between xt_jumpstack_alloc and xt_replace_table.

Indeed that seems to be so.

Quoted 21 lines
Quoted 6 lines
net/netfilter/x_tables.c +0 -0
index 445de70..47b1e79 100644

net/netfilter/x_tables.c struct xt_table *xt_register_table(struct net *net, · line 849

Quoted 10 lines
net/netfilter/x_tables.c +0 -4
@@ -844,10 +844,6 @@ struct xt_table *xt_register_table(struct net *net,
 	struct xt_table_info *private;
 	struct xt_table *t, *table;
 
-	ret = xt_jumpstack_alloc(newinfo);
-	if (ret < 0)
-		return ERR_PTR(ret);
-
 	/* Don't add one object to multiple lists. */
 	table = kmemdup(input_table, sizeof(struct xt_table), GFP_KERNEL);
 	if (!table) {

--
To unsubscribe from this list: send the line "unsubscribe netfilter" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#9 lore raw
Quoted 9 lines · Jan Engelhardt
Quoted 8 lines · Jan Engelhardt
Quoted 6 lines · Xiaotian Feng

In xt_register_table, xt_jumpstack_alloc is called first, later xt_replace_table is used. But in xt_replace_table, xt_jumpstack_alloc will be used again. Then the memory allocated by previous xt_jumpstack_alloc will be leaked. We can simply remove the previous xt_jumpstack_alloc because there aren't any users of newinfo between xt_jumpstack_alloc and xt_replace_table.

Quoted 3 lines
Quoted 1 line

Indeed that seems to be so.

Applied, thanks everyone.

#10 lore raw
In reply to Jan Engelhardt
Quoted 9 lines · Jan Engelhardt
Quoted 8 lines · Jan Engelhardt
Quoted 6 lines · Xiaotian Feng

In xt_register_table, xt_jumpstack_alloc is called first, later xt_replace_table is used. But in xt_replace_table, xt_jumpstack_alloc will be used again. Then the memory allocated by previous xt_jumpstack_alloc will be leaked. We can simply remove the previous xt_jumpstack_alloc because there aren't any users of newinfo between xt_jumpstack_alloc and xt_replace_table.

Quoted 3 lines
Quoted 1 line

Indeed that seems to be so.

Applied, thanks.