Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Alexey Dobriyan]

> -static inline int ebt_basic_match(struct ebt_entry *e, struct ethhdr *h,
> -   const struct net_device *in, const struct net_device *out)
> +static inline int
> +ebt_basic_match(const struct ebt_entry *e, const struct ethhdr *h,
> +                const struct net_device *in, const struct net_device *out)

These const modifiers are pointless because compilers are smart enough
to notice non-modifiability and generate the very same code in both cases.

Nowadays, half of functions declarations in generic
xtables/iptables/ip6tables/arptables
code are littered with const which makes them pretty unpleasant to read.

> -	struct ebt_entry *e = (struct ebt_entry *)chain->data;
> +	const struct ebt_entry *e = (struct ebt_entry *)chain->data;

And such things are wrong (not second const).

Please, at least, keep "struct net *" out of this.

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Patrick McHardy]

Alexey Dobriyan wrote:
>> -static inline int ebt_basic_match(struct ebt_entry *e, struct ethhdr *h,
>> -   const struct net_device *in, const struct net_device *out)
>> +static inline int
>> +ebt_basic_match(const struct ebt_entry *e, const struct ethhdr *h,
>> +                const struct net_device *in, const struct net_device *out)
> 
> These const modifiers are pointless because compilers are smart enough
> to notice non-modifiability and generate the very same code in both cases.
> 
> Nowadays, half of functions declarations in generic
> xtables/iptables/ip6tables/arptables
> code are littered with const which makes them pretty unpleasant to read.

Well, the benefit is that the compiler can catch accidental modification
of data that is supposed to be constant.

> 
>> -	struct ebt_entry *e = (struct ebt_entry *)chain->data;
>> +	const struct ebt_entry *e = (struct ebt_entry *)chain->data;
> 
> And such things are wrong (not second const).

What's wrong about this?

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Alexey Dobriyan]

On Thu, Feb 11, 2010 at 11:27 AM, Patrick McHardy <kaber@trash.net> wrote:
> Alexey Dobriyan wrote:
>>> -static inline int ebt_basic_match(struct ebt_entry *e, struct ethhdr *h,
>>> -   const struct net_device *in, const struct net_device *out)
>>> +static inline int
>>> +ebt_basic_match(const struct ebt_entry *e, const struct ethhdr *h,
>>> +                const struct net_device *in, const struct net_device *out)
>>
>> These const modifiers are pointless because compilers are smart enough
>> to notice non-modifiability and generate the very same code in both cases.
>>
>> Nowadays, half of functions declarations in generic
>> xtables/iptables/ip6tables/arptables
>> code are littered with const which makes them pretty unpleasant to read.
>
> Well, the benefit is that the compiler can catch accidental modification
> of data that is supposed to be constant.

Did it ever? Realistically?

From what I see, const is used with string functions
(to make string literals go to rodata) and with structures full of hooks
(to make them go to rodata).

>>> -    struct ebt_entry *e = (struct ebt_entry *)chain->data;
>>> +    const struct ebt_entry *e = (struct ebt_entry *)chain->data;
>>
>> And such things are wrong (not second const).
>
> What's wrong about this?

Should it be (const struct ebt_entry *)?
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Patrick McHardy]

Alexey Dobriyan wrote:
> On Thu, Feb 11, 2010 at 11:27 AM, Patrick McHardy <kaber@trash.net> wrote:
>> Alexey Dobriyan wrote:
>>>> -static inline int ebt_basic_match(struct ebt_entry *e, struct ethhdr *h,
>>>> -   const struct net_device *in, const struct net_device *out)
>>>> +static inline int
>>>> +ebt_basic_match(const struct ebt_entry *e, const struct ethhdr *h,
>>>> +                const struct net_device *in, const struct net_device *out)
>>> These const modifiers are pointless because compilers are smart enough
>>> to notice non-modifiability and generate the very same code in both cases.
>>>
>>> Nowadays, half of functions declarations in generic
>>> xtables/iptables/ip6tables/arptables
>>> code are littered with const which makes them pretty unpleasant to read.
>> Well, the benefit is that the compiler can catch accidental modification
>> of data that is supposed to be constant.
> 
> Did it ever? Realistically?

Yes, occasionally it has caught typos or other variable mixups for me.
It also documents that a variable is only supposed to be read.

>>From what I see, const is used with string functions
> (to make string literals go to rodata) and with structures full of hooks
> (to make them go to rodata).

Yes, that's also a reason.

>>>> -    struct ebt_entry *e = (struct ebt_entry *)chain->data;
>>>> +    const struct ebt_entry *e = (struct ebt_entry *)chain->data;
>>> And such things are wrong (not second const).
>> What's wrong about this?
> 
> Should it be (const struct ebt_entry *)?

That's not necessary. Its just declaring the the memory should
not be modified through e. The const qualification happens
automatically.

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Jan Engelhardt]

On Thursday 2010-02-11 10:56, Patrick McHardy wrote:

>Alexey Dobriyan wrote:
>> On Thu, Feb 11, 2010 at 11:27 AM, Patrick McHardy <kaber@trash.net> wrote:
>>> Alexey Dobriyan wrote:
>>>>> -static inline int ebt_basic_match(struct ebt_entry *e, struct ethhdr *h,
>>>>> -   const struct net_device *in, const struct net_device *out)
>>>>> +static inline int
>>>>> +ebt_basic_match(const struct ebt_entry *e, const struct ethhdr *h,
>>>>> +                const struct net_device *in, const struct net_device *out)
>>>> These const modifiers are pointless because compilers are smart enough
>>>> to notice non-modifiability and generate the very same code in both cases.
>
>Yes, occasionally it has caught typos or other variable mixups for me.
>It also documents that a variable is only supposed to be read.

In this case it allows parameter and local variable reduction
because I have proven, by means of const, that they would
remain the same over the entire function. (Scheduled commit
3c58069510:
http://dev.medozas.de/gitweb.cgi?p=linux;a=commit;h=3c58069510a498110240ceaff8c2a10c0d9b4460
 - this will be in the next pull group.)

>>>>> -    struct ebt_entry *e = (struct ebt_entry *)chain->data;
>>>>> +    const struct ebt_entry *e = (struct ebt_entry *)chain->data;
>>>> And such things are wrong (not second const).
>>> What's wrong about this?
>> 
>> Should it be (const struct ebt_entry *)?
>
>That's not necessary. Its just declaring the the memory should
>not be modified through e. The const qualification happens
>automatically.

Conversion from T* to const T* is permitted and implicit across =.

Code cleanups

[Jan Engelhardt]

Hi,


the next patch pack is mostly about cleanups, and there is one
commit for improving debug a little.
Pull URL will be posted once the previous request has been
merged, but I am already posting them now for review.



The following changes since commit e3eaa9910b380530cfd2c0670fcd3f627674da8a:
  Jan Engelhardt (1):
        netfilter: xtables: generate initial table on-demand

are available in the git repository at:

  . master2

Jan Engelhardt (6):
      netfilter: iptables: remove unused function arguments
      netfilter: reduce NF_HOOK by one argument
      netfilter: get rid of the grossness in netfilter.h
      netfilter: xtables: print details on size mismatch
      netfilter: xtables: constify args in compat copying functions
      netfilter: xtables: add const qualifiers

 include/linux/netfilter.h            |   55 +++++++++++---------
 include/linux/netfilter/x_tables.h   |   12 ++--
 net/bridge/netfilter/ebtables.c      |   59 ++++++++++++----------
 net/ipv4/netfilter/arp_tables.c      |   73 +++++++++++++++------------
 net/ipv4/netfilter/ip_tables.c       |   92 +++++++++++++++++++---------------
 net/ipv4/netfilter/ipt_ULOG.c        |    4 +-
 net/ipv4/netfilter/iptable_mangle.c  |   10 +---
 net/ipv6/netfilter/ip6_tables.c      |   92 ++++++++++++++++++---------------
 net/ipv6/netfilter/ip6table_mangle.c |   10 +---
 net/netfilter/x_tables.c             |   16 ++++--
 net/netfilter/xt_hashlimit.c         |    4 +-
 net/netfilter/xt_limit.c             |    4 +-
 12 files changed, 233 insertions(+), 198 deletions(-)

[PATCH 6/6] netfilter: xtables: add const qualifiers

[Jan Engelhardt]

This should make it easier to remove redundant arguments later.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
---
 net/bridge/netfilter/ebtables.c |   59 ++++++++++++++------------
 net/ipv4/netfilter/arp_tables.c |   69 +++++++++++++++++-------------
 net/ipv4/netfilter/ip_tables.c  |   88 +++++++++++++++++++++-----------------
 net/ipv6/netfilter/ip6_tables.c |   88 +++++++++++++++++++++------------------
 4 files changed, 168 insertions(+), 136 deletions(-)

diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c
index 208f4e3..bcdf02d 100644
--- a/net/bridge/netfilter/ebtables.c
+++ b/net/bridge/netfilter/ebtables.c
@@ -82,7 +82,8 @@ static inline int ebt_do_match (struct ebt_entry_match *m,
 	return m->u.match->match(skb, par) ? EBT_MATCH : EBT_NOMATCH;
 }
 
-static inline int ebt_dev_check(char *entry, const struct net_device *device)
+static inline int
+ebt_dev_check(const char *entry, const struct net_device *device)
 {
 	int i = 0;
 	const char *devname;
@@ -100,8 +101,9 @@ static inline int ebt_dev_check(char *entry, const struct net_device *device)
 
 #define FWINV2(bool,invflg) ((bool) ^ !!(e->invflags & invflg))
 /* process standard matches */
-static inline int ebt_basic_match(struct ebt_entry *e, struct ethhdr *h,
-   const struct net_device *in, const struct net_device *out)
+static inline int
+ebt_basic_match(const struct ebt_entry *e, const struct ethhdr *h,
+                const struct net_device *in, const struct net_device *out)
 {
 	int verdict, i;
 
@@ -156,12 +158,12 @@ unsigned int ebt_do_table (unsigned int hook, struct sk_buff *skb,
 	int i, nentries;
 	struct ebt_entry *point;
 	struct ebt_counter *counter_base, *cb_base;
-	struct ebt_entry_target *t;
+	const struct ebt_entry_target *t;
 	int verdict, sp = 0;
 	struct ebt_chainstack *cs;
 	struct ebt_entries *chaininfo;
-	char *base;
-	struct ebt_table_info *private;
+	const char *base;
+	const struct ebt_table_info *private;
 	bool hotdrop = false;
 	struct xt_match_param mtpar;
 	struct xt_target_param tgpar;
@@ -395,7 +397,7 @@ ebt_check_watcher(struct ebt_entry_watcher *w, struct xt_tgchk_param *par,
 	return 0;
 }
 
-static int ebt_verify_pointers(struct ebt_replace *repl,
+static int ebt_verify_pointers(const struct ebt_replace *repl,
 			       struct ebt_table_info *newinfo)
 {
 	unsigned int limit = repl->entries_size;
@@ -466,8 +468,8 @@ static int ebt_verify_pointers(struct ebt_replace *repl,
  * to parse the userspace data
  */
 static inline int
-ebt_check_entry_size_and_hooks(struct ebt_entry *e,
-   struct ebt_table_info *newinfo,
+ebt_check_entry_size_and_hooks(const struct ebt_entry *e,
+   const struct ebt_table_info *newinfo,
    unsigned int *n, unsigned int *cnt,
    unsigned int *totalcnt, unsigned int *udc_cnt)
 {
@@ -622,9 +624,8 @@ ebt_cleanup_entry(struct ebt_entry *e, struct net *net, unsigned int *cnt)
 }
 
 static inline int
-ebt_check_entry(struct ebt_entry *e,
-   struct net *net,
-   struct ebt_table_info *newinfo,
+ebt_check_entry(struct ebt_entry *e, struct net *net,
+   const struct ebt_table_info *newinfo,
    const char *name, unsigned int *cnt,
    struct ebt_cl_stack *cl_s, unsigned int udc_cnt)
 {
@@ -743,12 +744,12 @@ cleanup_matches:
  * the hook mask for udc tells us from which base chains the udc can be
  * accessed. This mask is a parameter to the check() functions of the extensions
  */
-static int check_chainloops(struct ebt_entries *chain, struct ebt_cl_stack *cl_s,
+static int check_chainloops(const struct ebt_entries *chain, struct ebt_cl_stack *cl_s,
    unsigned int udc_cnt, unsigned int hooknr, char *base)
 {
 	int i, chain_nr = -1, pos = 0, nentries = chain->nentries, verdict;
-	struct ebt_entry *e = (struct ebt_entry *)chain->data;
-	struct ebt_entry_target *t;
+	const struct ebt_entry *e = (struct ebt_entry *)chain->data;
+	const struct ebt_entry_target *t;
 
 	while (pos < nentries || chain_nr != -1) {
 		/* end of udc, go back one 'recursion' step */
@@ -814,7 +815,7 @@ letscontinue:
 }
 
 /* do the parsing of the table/chains/entries/matches/watchers/targets, heh */
-static int translate_table(struct net *net, char *name,
+static int translate_table(struct net *net, const char *name,
 			   struct ebt_table_info *newinfo)
 {
 	unsigned int i, j, k, udc_cnt;
@@ -934,7 +935,7 @@ static int translate_table(struct net *net, char *name,
 }
 
 /* called under write_lock */
-static void get_counters(struct ebt_counter *oldcounters,
+static void get_counters(const struct ebt_counter *oldcounters,
    struct ebt_counter *counters, unsigned int nentries)
 {
 	int i, cpu;
@@ -957,7 +958,8 @@ static void get_counters(struct ebt_counter *oldcounters,
 }
 
 /* replace the table */
-static int do_replace(struct net *net, void __user *user, unsigned int len)
+static int do_replace(struct net *net, const void __user *user,
+		      unsigned int len)
 {
 	int ret, i, countersize;
 	struct ebt_table_info *newinfo;
@@ -1237,7 +1239,8 @@ void ebt_unregister_table(struct net *net, struct ebt_table *table)
 }
 
 /* userspace just supplied us with counters */
-static int update_counters(struct net *net, void __user *user, unsigned int len)
+static int update_counters(struct net *net, const void __user *user,
+			   unsigned int len)
 {
 	int i, ret;
 	struct ebt_counter *tmp;
@@ -1292,8 +1295,8 @@ free_tmp:
 	return ret;
 }
 
-static inline int ebt_make_matchname(struct ebt_entry_match *m,
-   char *base, char __user *ubase)
+static inline int ebt_make_matchname(const struct ebt_entry_match *m,
+    const char *base, char __user *ubase)
 {
 	char __user *hlp = ubase + ((char *)m - base);
 	if (copy_to_user(hlp, m->u.match->name, EBT_FUNCTION_MAXNAMELEN))
@@ -1301,8 +1304,8 @@ static inline int ebt_make_matchname(struct ebt_entry_match *m,
 	return 0;
 }
 
-static inline int ebt_make_watchername(struct ebt_entry_watcher *w,
-   char *base, char __user *ubase)
+static inline int ebt_make_watchername(const struct ebt_entry_watcher *w,
+    const char *base, char __user *ubase)
 {
 	char __user *hlp = ubase + ((char *)w - base);
 	if (copy_to_user(hlp , w->u.watcher->name, EBT_FUNCTION_MAXNAMELEN))
@@ -1310,11 +1313,12 @@ static inline int ebt_make_watchername(struct ebt_entry_watcher *w,
 	return 0;
 }
 
-static inline int ebt_make_names(struct ebt_entry *e, char *base, char __user *ubase)
+static inline int
+ebt_make_names(struct ebt_entry *e, const char *base, char __user *ubase)
 {
 	int ret;
 	char __user *hlp;
-	struct ebt_entry_target *t;
+	const struct ebt_entry_target *t;
 
 	if (e->bitmask == 0)
 		return 0;
@@ -1335,10 +1339,11 @@ static inline int ebt_make_names(struct ebt_entry *e, char *base, char __user *u
 
 /* called with ebt_mutex locked */
 static int copy_everything_to_user(struct ebt_table *t, void __user *user,
-   int *len, int cmd)
+    const int *len, int cmd)
 {
 	struct ebt_replace tmp;
-	struct ebt_counter *counterstmp, *oldcounters;
+	struct ebt_counter *counterstmp;
+	const struct ebt_counter *oldcounters;
 	unsigned int entries_size, nentries;
 	char *entries;
 
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 2303dc9..4db5c1e 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -233,7 +233,14 @@ arpt_error(struct sk_buff *skb, const struct xt_target_param *par)
 	return NF_DROP;
 }
 
-static inline struct arpt_entry *get_entry(void *base, unsigned int offset)
+static inline const struct arpt_entry_target *
+arpt_get_target_c(const struct arpt_entry *e)
+{
+	return arpt_get_target((struct arpt_entry *)e);
+}
+
+static inline struct arpt_entry *
+get_entry(const void *base, unsigned int offset)
 {
 	return (struct arpt_entry *)(base + offset);
 }
@@ -280,7 +287,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 
 	arp = arp_hdr(skb);
 	do {
-		struct arpt_entry_target *t;
+		const struct arpt_entry_target *t;
 		int hdr_len;
 
 		if (!arp_packet_match(arp, skb->dev, indev, outdev, &e->arp)) {
@@ -292,7 +299,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 			(2 * skb->dev->addr_len);
 		ADD_COUNTER(e->counters, hdr_len, 1);
 
-		t = arpt_get_target(e);
+		t = arpt_get_target_c(e);
 
 		/* Standard target? */
 		if (!t->u.kernel.target->target) {
@@ -358,7 +365,7 @@ static inline bool unconditional(const struct arpt_arp *arp)
 /* Figures out from what hook each rule can be called: returns 0 if
  * there are loops.  Puts hook bitmask in comefrom.
  */
-static int mark_source_chains(struct xt_table_info *newinfo,
+static int mark_source_chains(const struct xt_table_info *newinfo,
 			      unsigned int valid_hooks, void *entry0)
 {
 	unsigned int hook;
@@ -379,7 +386,7 @@ static int mark_source_chains(struct xt_table_info *newinfo,
 
 		for (;;) {
 			const struct arpt_standard_target *t
-				= (void *)arpt_get_target(e);
+				= (void *)arpt_get_target_c(e);
 			int visited = e->comefrom & (1 << hook);
 
 			if (e->comefrom & (1 << NF_ARP_NUMHOOKS)) {
@@ -463,7 +470,7 @@ static int mark_source_chains(struct xt_table_info *newinfo,
 	return 1;
 }
 
-static inline int check_entry(struct arpt_entry *e, const char *name)
+static inline int check_entry(const struct arpt_entry *e, const char *name)
 {
 	const struct arpt_entry_target *t;
 
@@ -475,7 +482,7 @@ static inline int check_entry(struct arpt_entry *e, const char *name)
 	if (e->target_offset + sizeof(struct arpt_entry_target) > e->next_offset)
 		return -EINVAL;
 
-	t = arpt_get_target(e);
+	t = arpt_get_target_c(e);
 	if (e->target_offset + t->u.target_size > e->next_offset)
 		return -EINVAL;
 
@@ -540,14 +547,14 @@ out:
 	return ret;
 }
 
-static bool check_underflow(struct arpt_entry *e)
+static bool check_underflow(const struct arpt_entry *e)
 {
 	const struct arpt_entry_target *t;
 	unsigned int verdict;
 
 	if (!unconditional(&e->arp))
 		return false;
-	t = arpt_get_target(e);
+	t = arpt_get_target_c(e);
 	if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0)
 		return false;
 	verdict = ((struct arpt_standard_target *)t)->verdict;
@@ -557,8 +564,8 @@ static bool check_underflow(struct arpt_entry *e)
 
 static inline int check_entry_size_and_hooks(struct arpt_entry *e,
 					     struct xt_table_info *newinfo,
-					     unsigned char *base,
-					     unsigned char *limit,
+					     const unsigned char *base,
+					     const unsigned char *limit,
 					     const unsigned int *hook_entries,
 					     const unsigned int *underflows,
 					     unsigned int valid_hooks,
@@ -768,11 +775,11 @@ static void get_counters(const struct xt_table_info *t,
 	local_bh_enable();
 }
 
-static struct xt_counters *alloc_counters(struct xt_table *table)
+static struct xt_counters *alloc_counters(const struct xt_table *table)
 {
 	unsigned int countersize;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 
 	/* We need atomic snapshot of counters: rest doesn't change
 	 * (other than comefrom, which userspace doesn't care
@@ -790,11 +797,11 @@ static struct xt_counters *alloc_counters(struct xt_table *table)
 }
 
 static int copy_entries_to_user(unsigned int total_size,
-				struct xt_table *table,
+				const struct xt_table *table,
 				void __user *userptr)
 {
 	unsigned int off, num;
-	struct arpt_entry *e;
+	const struct arpt_entry *e;
 	struct xt_counters *counters;
 	struct xt_table_info *private = table->private;
 	int ret = 0;
@@ -814,7 +821,7 @@ static int copy_entries_to_user(unsigned int total_size,
 	/* FIXME: use iterator macros --RR */
 	/* ... then go back and fix counters and names */
 	for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){
-		struct arpt_entry_target *t;
+		const struct arpt_entry_target *t;
 
 		e = (struct arpt_entry *)(loc_cpu_entry + off);
 		if (copy_to_user(userptr + off
@@ -825,7 +832,7 @@ static int copy_entries_to_user(unsigned int total_size,
 			goto free_counters;
 		}
 
-		t = arpt_get_target(e);
+		t = arpt_get_target_c(e);
 		if (copy_to_user(userptr + off + e->target_offset
 				 + offsetof(struct arpt_entry_target,
 					    u.user.name),
@@ -860,18 +867,18 @@ static int compat_standard_to_user(void __user *dst, const void *src)
 	return copy_to_user(dst, &cv, sizeof(cv)) ? -EFAULT : 0;
 }
 
-static int compat_calc_entry(struct arpt_entry *e,
+static int compat_calc_entry(const struct arpt_entry *e,
 			     const struct xt_table_info *info,
-			     void *base, struct xt_table_info *newinfo)
+			     const void *base, struct xt_table_info *newinfo)
 {
-	struct arpt_entry_target *t;
+	const struct arpt_entry_target *t;
 	unsigned int entry_offset;
 	int off, i, ret;
 
 	off = sizeof(struct arpt_entry) - sizeof(struct compat_arpt_entry);
 	entry_offset = (void *)e - base;
 
-	t = arpt_get_target(e);
+	t = arpt_get_target_c(e);
 	off += xt_compat_target_offset(t->u.kernel.target);
 	newinfo->size -= off;
 	ret = xt_compat_add_offset(NFPROTO_ARP, entry_offset, off);
@@ -907,7 +914,8 @@ static int compat_table_info(const struct xt_table_info *info,
 }
 #endif
 
-static int get_info(struct net *net, void __user *user, int *len, int compat)
+static int get_info(struct net *net, void __user *user,
+                    const int *len, int compat)
 {
 	char name[ARPT_TABLE_MAXNAMELEN];
 	struct xt_table *t;
@@ -966,7 +974,7 @@ static int get_info(struct net *net, void __user *user, int *len, int compat)
 }
 
 static int get_entries(struct net *net, struct arpt_get_entries __user *uptr,
-		       int *len)
+		       const int *len)
 {
 	int ret;
 	struct arpt_get_entries get;
@@ -1080,7 +1088,8 @@ static int __do_replace(struct net *net, const char *name,
 	return ret;
 }
 
-static int do_replace(struct net *net, void __user *user, unsigned int len)
+static int do_replace(struct net *net, const void __user *user,
+                      unsigned int len)
 {
 	int ret;
 	struct arpt_replace tmp;
@@ -1140,8 +1149,8 @@ add_counter_to_entry(struct arpt_entry *e,
 	return 0;
 }
 
-static int do_add_counters(struct net *net, void __user *user, unsigned int len,
-			   int compat)
+static int do_add_counters(struct net *net, const void __user *user,
+			   unsigned int len, int compat)
 {
 	unsigned int i, curcpu;
 	struct xt_counters_info tmp;
@@ -1245,10 +1254,10 @@ static inline int
 check_compat_entry_size_and_hooks(struct compat_arpt_entry *e,
 				  struct xt_table_info *newinfo,
 				  unsigned int *size,
-				  unsigned char *base,
-				  unsigned char *limit,
-				  unsigned int *hook_entries,
-				  unsigned int *underflows,
+				  const unsigned char *base,
+				  const unsigned char *limit,
+				  const unsigned int *hook_entries,
+				  const unsigned int *underflows,
 				  unsigned int *i,
 				  const char *name)
 {
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 2a4f745..e94c18b 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -176,7 +176,7 @@ ipt_error(struct sk_buff *skb, const struct xt_target_param *par)
 
 /* Performance critical - called for every packet */
 static inline bool
-do_match(struct ipt_entry_match *m, const struct sk_buff *skb,
+do_match(const struct ipt_entry_match *m, const struct sk_buff *skb,
 	 struct xt_match_param *par)
 {
 	par->match     = m->u.kernel.match;
@@ -191,7 +191,7 @@ do_match(struct ipt_entry_match *m, const struct sk_buff *skb,
 
 /* Performance critical */
 static inline struct ipt_entry *
-get_entry(void *base, unsigned int offset)
+get_entry(const void *base, unsigned int offset)
 {
 	return (struct ipt_entry *)(base + offset);
 }
@@ -206,6 +206,13 @@ static inline bool unconditional(const struct ipt_ip *ip)
 #undef FWINV
 }
 
+/* for const-correctness */
+static inline const struct ipt_entry_target *
+ipt_get_target_c(const struct ipt_entry *e)
+{
+	return ipt_get_target((struct ipt_entry *)e);
+}
+
 #if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
     defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
 static const char *const hooknames[] = {
@@ -240,11 +247,11 @@ static struct nf_loginfo trace_loginfo = {
 
 /* Mildly perf critical (only if packet tracing is on) */
 static inline int
-get_chainname_rulenum(struct ipt_entry *s, struct ipt_entry *e,
+get_chainname_rulenum(const struct ipt_entry *s, const struct ipt_entry *e,
 		      const char *hookname, const char **chainname,
 		      const char **comment, unsigned int *rulenum)
 {
-	struct ipt_standard_target *t = (void *)ipt_get_target(s);
+	const struct ipt_standard_target *t = (void *)ipt_get_target_c(s);
 
 	if (strcmp(t->target.u.kernel.target->name, IPT_ERROR_TARGET) == 0) {
 		/* Head of user chain: ERROR target with chainname */
@@ -270,15 +277,15 @@ get_chainname_rulenum(struct ipt_entry *s, struct ipt_entry *e,
 	return 0;
 }
 
-static void trace_packet(struct sk_buff *skb,
+static void trace_packet(const struct sk_buff *skb,
 			 unsigned int hook,
 			 const struct net_device *in,
 			 const struct net_device *out,
 			 const char *tablename,
-			 struct xt_table_info *private,
-			 struct ipt_entry *e)
+			 const struct xt_table_info *private,
+			 const struct ipt_entry *e)
 {
-	void *table_base;
+	const void *table_base;
 	const struct ipt_entry *root;
 	const char *hookname, *chainname, *comment;
 	unsigned int rulenum = 0;
@@ -322,9 +329,9 @@ ipt_do_table(struct sk_buff *skb,
 	/* Initializing verdict to NF_DROP keeps gcc happy. */
 	unsigned int verdict = NF_DROP;
 	const char *indev, *outdev;
-	void *table_base;
+	const void *table_base;
 	struct ipt_entry *e, *back;
-	struct xt_table_info *private;
+	const struct xt_table_info *private;
 	struct xt_match_param mtpar;
 	struct xt_target_param tgpar;
 
@@ -357,7 +364,7 @@ ipt_do_table(struct sk_buff *skb,
 	back = get_entry(table_base, private->underflow[hook]);
 
 	do {
-		struct ipt_entry_target *t;
+		const struct ipt_entry_target *t;
 
 		IP_NF_ASSERT(e);
 		IP_NF_ASSERT(back);
@@ -450,7 +457,7 @@ ipt_do_table(struct sk_buff *skb,
 /* Figures out from what hook each rule can be called: returns 0 if
    there are loops.  Puts hook bitmask in comefrom. */
 static int
-mark_source_chains(struct xt_table_info *newinfo,
+mark_source_chains(const struct xt_table_info *newinfo,
 		   unsigned int valid_hooks, void *entry0)
 {
 	unsigned int hook;
@@ -468,8 +475,8 @@ mark_source_chains(struct xt_table_info *newinfo,
 		e->counters.pcnt = pos;
 
 		for (;;) {
-			struct ipt_standard_target *t
-				= (void *)ipt_get_target(e);
+			const struct ipt_standard_target *t
+				= (void *)ipt_get_target_c(e);
 			int visited = e->comefrom & (1 << hook);
 
 			if (e->comefrom & (1 << NF_INET_NUMHOOKS)) {
@@ -578,9 +585,9 @@ cleanup_match(struct ipt_entry_match *m, struct net *net, unsigned int *i)
 }
 
 static int
-check_entry(struct ipt_entry *e, const char *name)
+check_entry(const struct ipt_entry *e, const char *name)
 {
-	struct ipt_entry_target *t;
+	const struct ipt_entry_target *t;
 
 	if (!ip_checkentry(&e->ip)) {
 		duprintf("ip_tables: ip check failed %p %s.\n", e, name);
@@ -591,7 +598,7 @@ check_entry(struct ipt_entry *e, const char *name)
 	    e->next_offset)
 		return -EINVAL;
 
-	t = ipt_get_target(e);
+	t = ipt_get_target_c(e);
 	if (e->target_offset + t->u.target_size > e->next_offset)
 		return -EINVAL;
 
@@ -718,14 +725,14 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name,
 	return ret;
 }
 
-static bool check_underflow(struct ipt_entry *e)
+static bool check_underflow(const struct ipt_entry *e)
 {
 	const struct ipt_entry_target *t;
 	unsigned int verdict;
 
 	if (!unconditional(&e->ip))
 		return false;
-	t = ipt_get_target(e);
+	t = ipt_get_target_c(e);
 	if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0)
 		return false;
 	verdict = ((struct ipt_standard_target *)t)->verdict;
@@ -736,8 +743,8 @@ static bool check_underflow(struct ipt_entry *e)
 static int
 check_entry_size_and_hooks(struct ipt_entry *e,
 			   struct xt_table_info *newinfo,
-			   unsigned char *base,
-			   unsigned char *limit,
+			   const unsigned char *base,
+			   const unsigned char *limit,
 			   const unsigned int *hook_entries,
 			   const unsigned int *underflows,
 			   unsigned int valid_hooks,
@@ -952,11 +959,11 @@ get_counters(const struct xt_table_info *t,
 	local_bh_enable();
 }
 
-static struct xt_counters * alloc_counters(struct xt_table *table)
+static struct xt_counters *alloc_counters(const struct xt_table *table)
 {
 	unsigned int countersize;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 
 	/* We need atomic snapshot of counters: rest doesn't change
 	   (other than comefrom, which userspace doesn't care
@@ -974,11 +981,11 @@ static struct xt_counters * alloc_counters(struct xt_table *table)
 
 static int
 copy_entries_to_user(unsigned int total_size,
-		     struct xt_table *table,
+		     const struct xt_table *table,
 		     void __user *userptr)
 {
 	unsigned int off, num;
-	struct ipt_entry *e;
+	const struct ipt_entry *e;
 	struct xt_counters *counters;
 	const struct xt_table_info *private = table->private;
 	int ret = 0;
@@ -1030,7 +1037,7 @@ copy_entries_to_user(unsigned int total_size,
 			}
 		}
 
-		t = ipt_get_target(e);
+		t = ipt_get_target_c(e);
 		if (copy_to_user(userptr + off + e->target_offset
 				 + offsetof(struct ipt_entry_target,
 					    u.user.name),
@@ -1066,24 +1073,24 @@ static int compat_standard_to_user(void __user *dst, const void *src)
 }
 
 static inline int
-compat_calc_match(struct ipt_entry_match *m, int *size)
+compat_calc_match(const struct ipt_entry_match *m, int *size)
 {
 	*size += xt_compat_match_offset(m->u.kernel.match);
 	return 0;
 }
 
-static int compat_calc_entry(struct ipt_entry *e,
+static int compat_calc_entry(const struct ipt_entry *e,
 			     const struct xt_table_info *info,
-			     void *base, struct xt_table_info *newinfo)
+			     const void *base, struct xt_table_info *newinfo)
 {
-	struct ipt_entry_target *t;
+	const struct ipt_entry_target *t;
 	unsigned int entry_offset;
 	int off, i, ret;
 
 	off = sizeof(struct ipt_entry) - sizeof(struct compat_ipt_entry);
 	entry_offset = (void *)e - base;
 	IPT_MATCH_ITERATE(e, compat_calc_match, &off);
-	t = ipt_get_target(e);
+	t = ipt_get_target_c(e);
 	off += xt_compat_target_offset(t->u.kernel.target);
 	newinfo->size -= off;
 	ret = xt_compat_add_offset(AF_INET, entry_offset, off);
@@ -1119,7 +1126,8 @@ static int compat_table_info(const struct xt_table_info *info,
 }
 #endif
 
-static int get_info(struct net *net, void __user *user, int *len, int compat)
+static int get_info(struct net *net, void __user *user,
+                    const int *len, int compat)
 {
 	char name[IPT_TABLE_MAXNAMELEN];
 	struct xt_table *t;
@@ -1179,7 +1187,8 @@ static int get_info(struct net *net, void __user *user, int *len, int compat)
 }
 
 static int
-get_entries(struct net *net, struct ipt_get_entries __user *uptr, int *len)
+get_entries(struct net *net, struct ipt_get_entries __user *uptr,
+	    const int *len)
 {
 	int ret;
 	struct ipt_get_entries get;
@@ -1289,7 +1298,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks,
 }
 
 static int
-do_replace(struct net *net, void __user *user, unsigned int len)
+do_replace(struct net *net, const void __user *user, unsigned int len)
 {
 	int ret;
 	struct ipt_replace tmp;
@@ -1350,7 +1359,8 @@ add_counter_to_entry(struct ipt_entry *e,
 }
 
 static int
-do_add_counters(struct net *net, void __user *user, unsigned int len, int compat)
+do_add_counters(struct net *net, const void __user *user,
+                unsigned int len, int compat)
 {
 	unsigned int i, curcpu;
 	struct xt_counters_info tmp;
@@ -1546,10 +1556,10 @@ static int
 check_compat_entry_size_and_hooks(struct compat_ipt_entry *e,
 				  struct xt_table_info *newinfo,
 				  unsigned int *size,
-				  unsigned char *base,
-				  unsigned char *limit,
-				  unsigned int *hook_entries,
-				  unsigned int *underflows,
+				  const unsigned char *base,
+				  const unsigned char *limit,
+				  const unsigned int *hook_entries,
+				  const unsigned int *underflows,
 				  unsigned int *i,
 				  const char *name)
 {
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 3ff4fd5..4185099 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -208,7 +208,7 @@ ip6t_error(struct sk_buff *skb, const struct xt_target_param *par)
 
 /* Performance critical - called for every packet */
 static inline bool
-do_match(struct ip6t_entry_match *m, const struct sk_buff *skb,
+do_match(const struct ip6t_entry_match *m, const struct sk_buff *skb,
 	 struct xt_match_param *par)
 {
 	par->match     = m->u.kernel.match;
@@ -222,7 +222,7 @@ do_match(struct ip6t_entry_match *m, const struct sk_buff *skb,
 }
 
 static inline struct ip6t_entry *
-get_entry(void *base, unsigned int offset)
+get_entry(const void *base, unsigned int offset)
 {
 	return (struct ip6t_entry *)(base + offset);
 }
@@ -236,6 +236,12 @@ static inline bool unconditional(const struct ip6t_ip6 *ipv6)
 	return memcmp(ipv6, &uncond, sizeof(uncond)) == 0;
 }
 
+static inline const struct ip6t_entry_target *
+ip6t_get_target_c(const struct ip6t_entry *e)
+{
+	return ip6t_get_target((struct ip6t_entry *)e);
+}
+
 #if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
     defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
 /* This cries for unification! */
@@ -271,11 +277,11 @@ static struct nf_loginfo trace_loginfo = {
 
 /* Mildly perf critical (only if packet tracing is on) */
 static inline int
-get_chainname_rulenum(struct ip6t_entry *s, struct ip6t_entry *e,
+get_chainname_rulenum(const struct ip6t_entry *s, const struct ip6t_entry *e,
 		      const char *hookname, const char **chainname,
 		      const char **comment, unsigned int *rulenum)
 {
-	struct ip6t_standard_target *t = (void *)ip6t_get_target(s);
+	const struct ip6t_standard_target *t = (void *)ip6t_get_target_c(s);
 
 	if (strcmp(t->target.u.kernel.target->name, IP6T_ERROR_TARGET) == 0) {
 		/* Head of user chain: ERROR target with chainname */
@@ -301,15 +307,15 @@ get_chainname_rulenum(struct ip6t_entry *s, struct ip6t_entry *e,
 	return 0;
 }
 
-static void trace_packet(struct sk_buff *skb,
+static void trace_packet(const struct sk_buff *skb,
 			 unsigned int hook,
 			 const struct net_device *in,
 			 const struct net_device *out,
 			 const char *tablename,
-			 struct xt_table_info *private,
-			 struct ip6t_entry *e)
+			 const struct xt_table_info *private,
+			 const struct ip6t_entry *e)
 {
-	void *table_base;
+	const void *table_base;
 	const struct ip6t_entry *root;
 	const char *hookname, *chainname, *comment;
 	unsigned int rulenum = 0;
@@ -352,9 +358,9 @@ ip6t_do_table(struct sk_buff *skb,
 	/* Initializing verdict to NF_DROP keeps gcc happy. */
 	unsigned int verdict = NF_DROP;
 	const char *indev, *outdev;
-	void *table_base;
+	const void *table_base;
 	struct ip6t_entry *e, *back;
-	struct xt_table_info *private;
+	const struct xt_table_info *private;
 	struct xt_match_param mtpar;
 	struct xt_target_param tgpar;
 
@@ -385,7 +391,7 @@ ip6t_do_table(struct sk_buff *skb,
 	back = get_entry(table_base, private->underflow[hook]);
 
 	do {
-		struct ip6t_entry_target *t;
+		const struct ip6t_entry_target *t;
 
 		IP_NF_ASSERT(e);
 		IP_NF_ASSERT(back);
@@ -400,7 +406,7 @@ ip6t_do_table(struct sk_buff *skb,
 			    ntohs(ipv6_hdr(skb)->payload_len) +
 			    sizeof(struct ipv6hdr), 1);
 
-		t = ip6t_get_target(e);
+		t = ip6t_get_target_c(e);
 		IP_NF_ASSERT(t->u.kernel.target);
 
 #if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
@@ -482,7 +488,7 @@ ip6t_do_table(struct sk_buff *skb,
 /* Figures out from what hook each rule can be called: returns 0 if
    there are loops.  Puts hook bitmask in comefrom. */
 static int
-mark_source_chains(struct xt_table_info *newinfo,
+mark_source_chains(const struct xt_table_info *newinfo,
 		   unsigned int valid_hooks, void *entry0)
 {
 	unsigned int hook;
@@ -500,8 +506,8 @@ mark_source_chains(struct xt_table_info *newinfo,
 		e->counters.pcnt = pos;
 
 		for (;;) {
-			struct ip6t_standard_target *t
-				= (void *)ip6t_get_target(e);
+			const struct ip6t_standard_target *t
+				= (void *)ip6t_get_target_c(e);
 			int visited = e->comefrom & (1 << hook);
 
 			if (e->comefrom & (1 << NF_INET_NUMHOOKS)) {
@@ -610,9 +616,9 @@ cleanup_match(struct ip6t_entry_match *m, struct net *net, unsigned int *i)
 }
 
 static int
-check_entry(struct ip6t_entry *e, const char *name)
+check_entry(const struct ip6t_entry *e, const char *name)
 {
-	struct ip6t_entry_target *t;
+	const struct ip6t_entry_target *t;
 
 	if (!ip6_checkentry(&e->ipv6)) {
 		duprintf("ip_tables: ip check failed %p %s.\n", e, name);
@@ -623,7 +629,7 @@ check_entry(struct ip6t_entry *e, const char *name)
 	    e->next_offset)
 		return -EINVAL;
 
-	t = ip6t_get_target(e);
+	t = ip6t_get_target_c(e);
 	if (e->target_offset + t->u.target_size > e->next_offset)
 		return -EINVAL;
 
@@ -750,14 +756,14 @@ find_check_entry(struct ip6t_entry *e, struct net *net, const char *name,
 	return ret;
 }
 
-static bool check_underflow(struct ip6t_entry *e)
+static bool check_underflow(const struct ip6t_entry *e)
 {
 	const struct ip6t_entry_target *t;
 	unsigned int verdict;
 
 	if (!unconditional(&e->ipv6))
 		return false;
-	t = ip6t_get_target(e);
+	t = ip6t_get_target_c(e);
 	if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0)
 		return false;
 	verdict = ((struct ip6t_standard_target *)t)->verdict;
@@ -768,8 +774,8 @@ static bool check_underflow(struct ip6t_entry *e)
 static int
 check_entry_size_and_hooks(struct ip6t_entry *e,
 			   struct xt_table_info *newinfo,
-			   unsigned char *base,
-			   unsigned char *limit,
+			   const unsigned char *base,
+			   const unsigned char *limit,
 			   const unsigned int *hook_entries,
 			   const unsigned int *underflows,
 			   unsigned int valid_hooks,
@@ -984,11 +990,11 @@ get_counters(const struct xt_table_info *t,
 	local_bh_enable();
 }
 
-static struct xt_counters *alloc_counters(struct xt_table *table)
+static struct xt_counters *alloc_counters(const struct xt_table *table)
 {
 	unsigned int countersize;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 
 	/* We need atomic snapshot of counters: rest doesn't change
 	   (other than comefrom, which userspace doesn't care
@@ -1006,11 +1012,11 @@ static struct xt_counters *alloc_counters(struct xt_table *table)
 
 static int
 copy_entries_to_user(unsigned int total_size,
-		     struct xt_table *table,
+		     const struct xt_table *table,
 		     void __user *userptr)
 {
 	unsigned int off, num;
-	struct ip6t_entry *e;
+	const struct ip6t_entry *e;
 	struct xt_counters *counters;
 	const struct xt_table_info *private = table->private;
 	int ret = 0;
@@ -1062,7 +1068,7 @@ copy_entries_to_user(unsigned int total_size,
 			}
 		}
 
-		t = ip6t_get_target(e);
+		t = ip6t_get_target_c(e);
 		if (copy_to_user(userptr + off + e->target_offset
 				 + offsetof(struct ip6t_entry_target,
 					    u.user.name),
@@ -1098,24 +1104,24 @@ static int compat_standard_to_user(void __user *dst, const void *src)
 }
 
 static inline int
-compat_calc_match(struct ip6t_entry_match *m, int *size)
+compat_calc_match(const struct ip6t_entry_match *m, int *size)
 {
 	*size += xt_compat_match_offset(m->u.kernel.match);
 	return 0;
 }
 
-static int compat_calc_entry(struct ip6t_entry *e,
+static int compat_calc_entry(const struct ip6t_entry *e,
 			     const struct xt_table_info *info,
-			     void *base, struct xt_table_info *newinfo)
+			     const void *base, struct xt_table_info *newinfo)
 {
-	struct ip6t_entry_target *t;
+	const struct ip6t_entry_target *t;
 	unsigned int entry_offset;
 	int off, i, ret;
 
 	off = sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry);
 	entry_offset = (void *)e - base;
 	IP6T_MATCH_ITERATE(e, compat_calc_match, &off);
-	t = ip6t_get_target(e);
+	t = ip6t_get_target_c(e);
 	off += xt_compat_target_offset(t->u.kernel.target);
 	newinfo->size -= off;
 	ret = xt_compat_add_offset(AF_INET6, entry_offset, off);
@@ -1151,7 +1157,8 @@ static int compat_table_info(const struct xt_table_info *info,
 }
 #endif
 
-static int get_info(struct net *net, void __user *user, int *len, int compat)
+static int get_info(struct net *net, void __user *user,
+                    const int *len, int compat)
 {
 	char name[IP6T_TABLE_MAXNAMELEN];
 	struct xt_table *t;
@@ -1211,7 +1218,8 @@ static int get_info(struct net *net, void __user *user, int *len, int compat)
 }
 
 static int
-get_entries(struct net *net, struct ip6t_get_entries __user *uptr, int *len)
+get_entries(struct net *net, struct ip6t_get_entries __user *uptr,
+            const int *len)
 {
 	int ret;
 	struct ip6t_get_entries get;
@@ -1322,7 +1330,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks,
 }
 
 static int
-do_replace(struct net *net, void __user *user, unsigned int len)
+do_replace(struct net *net, const void __user *user, unsigned int len)
 {
 	int ret;
 	struct ip6t_replace tmp;
@@ -1383,7 +1391,7 @@ add_counter_to_entry(struct ip6t_entry *e,
 }
 
 static int
-do_add_counters(struct net *net, void __user *user, unsigned int len,
+do_add_counters(struct net *net, const void __user *user, unsigned int len,
 		int compat)
 {
 	unsigned int i, curcpu;
@@ -1582,10 +1590,10 @@ static int
 check_compat_entry_size_and_hooks(struct compat_ip6t_entry *e,
 				  struct xt_table_info *newinfo,
 				  unsigned int *size,
-				  unsigned char *base,
-				  unsigned char *limit,
-				  unsigned int *hook_entries,
-				  unsigned int *underflows,
+				  const unsigned char *base,
+				  const unsigned char *limit,
+				  const unsigned int *hook_entries,
+				  const unsigned int *underflows,
 				  unsigned int *i,
 				  const char *name)
 {
-- 
1.6.6.1

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Patrick McHardy]

Jan Engelhardt wrote:
> +static inline const struct ip6t_entry_target *
> +ip6t_get_target_c(const struct ip6t_entry *e)
> +{
> +	return ip6t_get_target((struct ip6t_entry *)e);
> +}
> +

I would prefer a macro over these get_target_c functions.

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Jan Engelhardt]

On Thursday 2010-02-11 10:00, Patrick McHardy wrote:

>Jan Engelhardt wrote:
>> +static inline const struct ip6t_entry_target *
>> +ip6t_get_target_c(const struct ip6t_entry *e)
>> +{
>> +	return ip6t_get_target((struct ip6t_entry *)e);
>> +}
>> +
>
>I would prefer a macro over these get_target_c functions.

Why so?

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Patrick McHardy]

Jan Engelhardt wrote:
> On Thursday 2010-02-11 10:00, Patrick McHardy wrote:
> 
>> Jan Engelhardt wrote:
>>> +static inline const struct ip6t_entry_target *
>>> +ip6t_get_target_c(const struct ip6t_entry *e)
>>> +{
>>> +	return ip6t_get_target((struct ip6t_entry *)e);
>>> +}
>>> +
>> I would prefer a macro over these get_target_c functions.
> 
> Why so?

Because it avoids having a function for const and one for non-const.

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Jan Engelhardt]

On Thursday 2010-02-11 17:06, Patrick McHardy wrote:

>Jan Engelhardt wrote:
>> On Thursday 2010-02-11 10:00, Patrick McHardy wrote:
>> 
>>> Jan Engelhardt wrote:
>>>> +static inline const struct ip6t_entry_target *
>>>> +ip6t_get_target_c(const struct ip6t_entry *e)
>>>> +{
>>>> +	return ip6t_get_target((struct ip6t_entry *)e);
>>>> +}
>>>> +
>>> I would prefer a macro over these get_target_c functions.
>> 
>> Why so?
>
>Because it avoids having a function for const and one for non-const.

Well, I need two. One that returns const and one that does not;
ip6t_get_target_c is signaturally-incompatible with ip6t_get_target.
FWIW, the function is removed later on anyway ("remove remaining xt1 
code").

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Patrick McHardy]

Jan Engelhardt wrote:
> On Thursday 2010-02-11 17:06, Patrick McHardy wrote:
> 
>> Jan Engelhardt wrote:
>>> On Thursday 2010-02-11 10:00, Patrick McHardy wrote:
>>>
>>>> Jan Engelhardt wrote:
>>>>> +static inline const struct ip6t_entry_target *
>>>>> +ip6t_get_target_c(const struct ip6t_entry *e)
>>>>> +{
>>>>> +	return ip6t_get_target((struct ip6t_entry *)e);
>>>>> +}
>>>>> +
>>>> I would prefer a macro over these get_target_c functions.
>>> Why so?
>> Because it avoids having a function for const and one for non-const.
> 
> Well, I need two. One that returns const and one that does not;
> ip6t_get_target_c is signaturally-incompatible with ip6t_get_target.

Yes, but you wouldn't if you'd use a macro.

#define ip6t_get_target(e)	((void *)e + e->target_offset)

> FWIW, the function is removed later on anyway ("remove remaining xt1 
> code").

This is patch 6/6, what do you mean with "later"?

Re: [PATCH 6/6] netfilter: xtables: add const qualifiers

[Jan Engelhardt]

On Thursday 2010-02-11 17:27, Patrick McHardy wrote:
>>>>> I would prefer a macro over these get_target_c functions.
>>>> Why so?
>>> Because it avoids having a function for const and one for non-const.
>> 
>> Well, I need two. One that returns const and one that does not;
>> ip6t_get_target_c is signaturally-incompatible with ip6t_get_target.
>
>Yes, but you wouldn't if you'd use a macro.
>
>#define ip6t_get_target(e)	((void *)e + e->target_offset)

That would remove the const from a const e, which is what I
expressly wanted to avoid because it's already a little hairy.

>> FWIW, the function is removed later on anyway ("remove remaining xt1 
>> code").
>
>This is patch 6/6, what do you mean with "later"?

Patch 094/103 from the original submission
http://markmail.org/browse/org.kernel.vger.netfilter-devel/2009-08
that I am trying to push to you all the time :)

Xtables cleanup patches

[Jan Engelhardt]

I am posting here already the next round of pullable patches
(that I originally wanted to get in _much_ faster) for review.

This block is mostly cleanups, simplifications. It might be worth
testing these together with the previous set if there are concerns
about speed.

-----
The following changes since commit 7891b6c280d01d52be5ecefe0db79d4b1f671a1e:
  Jan Engelhardt (1):
        netfilter: xtables: generate initial table on-demand

are available in the git repository at:

  git://dev.medozas.de/linux next

Jan Engelhardt (6):
      netfilter: iptables: remove unused function arguments
      netfilter: reduce NF_HOOK by one argument
      netfilter: get rid of the grossness in netfilter.h
      netfilter: xtables: print details on size mismatch
      netfilter: xtables: constify args in compat copying functions
      netfilter: xtables: add const qualifiers

 include/linux/netfilter.h            |   55 +++++++++++---------
 include/linux/netfilter/x_tables.h   |   12 ++--
 net/bridge/netfilter/ebtables.c      |   56 +++++++++++---------
 net/ipv4/netfilter/arp_tables.c      |   73 +++++++++++++++------------
 net/ipv4/netfilter/ip_tables.c       |   92 +++++++++++++++++++---------------
 net/ipv4/netfilter/ipt_ULOG.c        |    4 +-
 net/ipv4/netfilter/iptable_mangle.c  |   10 +---
 net/ipv6/netfilter/ip6_tables.c      |   92 ++++++++++++++++++---------------
 net/ipv6/netfilter/ip6table_mangle.c |   10 +---
 net/netfilter/x_tables.c             |   16 ++++--
 net/netfilter/xt_hashlimit.c         |    4 +-
 net/netfilter/xt_limit.c             |    4 +-
 12 files changed, 232 insertions(+), 196 deletions(-)

[PATCH 6/6] netfilter: xtables: add const qualifiers

[Jan Engelhardt]

This should make it easier to remove redundant arguments later.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
---
 net/bridge/netfilter/ebtables.c |   56 ++++++++++++++-----------
 net/ipv4/netfilter/arp_tables.c |   69 +++++++++++++++++-------------
 net/ipv4/netfilter/ip_tables.c  |   88 +++++++++++++++++++++-----------------
 net/ipv6/netfilter/ip6_tables.c |   88 +++++++++++++++++++++------------------
 4 files changed, 167 insertions(+), 134 deletions(-)

diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c
index bd1c654..9ed98d9 100644
--- a/net/bridge/netfilter/ebtables.c
+++ b/net/bridge/netfilter/ebtables.c
@@ -82,7 +82,8 @@ static inline int ebt_do_match (struct ebt_entry_match *m,
 	return m->u.match->match(skb, par) ? EBT_MATCH : EBT_NOMATCH;
 }
 
-static inline int ebt_dev_check(char *entry, const struct net_device *device)
+static inline int
+ebt_dev_check(const char *entry, const struct net_device *device)
 {
 	int i = 0;
 	const char *devname;
@@ -100,8 +101,9 @@ static inline int ebt_dev_check(char *entry, const struct net_device *device)
 
 #define FWINV2(bool,invflg) ((bool) ^ !!(e->invflags & invflg))
 /* process standard matches */
-static inline int ebt_basic_match(struct ebt_entry *e, struct ethhdr *h,
-   const struct net_device *in, const struct net_device *out)
+static inline int
+ebt_basic_match(const struct ebt_entry *e, const struct ethhdr *h,
+                const struct net_device *in, const struct net_device *out)
 {
 	int verdict, i;
 
@@ -156,12 +158,12 @@ unsigned int ebt_do_table (unsigned int hook, struct sk_buff *skb,
 	int i, nentries;
 	struct ebt_entry *point;
 	struct ebt_counter *counter_base, *cb_base;
-	struct ebt_entry_target *t;
+	const struct ebt_entry_target *t;
 	int verdict, sp = 0;
 	struct ebt_chainstack *cs;
 	struct ebt_entries *chaininfo;
-	char *base;
-	struct ebt_table_info *private;
+	const char *base;
+	const struct ebt_table_info *private;
 	bool hotdrop = false;
 	struct xt_match_param mtpar;
 	struct xt_target_param tgpar;
@@ -395,7 +397,7 @@ ebt_check_watcher(struct ebt_entry_watcher *w, struct xt_tgchk_param *par,
 	return 0;
 }
 
-static int ebt_verify_pointers(struct ebt_replace *repl,
+static int ebt_verify_pointers(const struct ebt_replace *repl,
 			       struct ebt_table_info *newinfo)
 {
 	unsigned int limit = repl->entries_size;
@@ -466,8 +468,8 @@ static int ebt_verify_pointers(struct ebt_replace *repl,
  * to parse the userspace data
  */
 static inline int
-ebt_check_entry_size_and_hooks(struct ebt_entry *e,
-   struct ebt_table_info *newinfo,
+ebt_check_entry_size_and_hooks(const struct ebt_entry *e,
+   const struct ebt_table_info *newinfo,
    unsigned int *n, unsigned int *cnt,
    unsigned int *totalcnt, unsigned int *udc_cnt)
 {
@@ -619,7 +621,7 @@ ebt_cleanup_entry(struct ebt_entry *e, unsigned int *cnt)
 }
 
 static inline int
-ebt_check_entry(struct ebt_entry *e, struct ebt_table_info *newinfo,
+ebt_check_entry(struct ebt_entry *e, const struct ebt_table_info *newinfo,
    const char *name, unsigned int *cnt,
    struct ebt_cl_stack *cl_s, unsigned int udc_cnt)
 {
@@ -737,12 +739,12 @@ cleanup_matches:
  * the hook mask for udc tells us from which base chains the udc can be
  * accessed. This mask is a parameter to the check() functions of the extensions
  */
-static int check_chainloops(struct ebt_entries *chain, struct ebt_cl_stack *cl_s,
+static int check_chainloops(const struct ebt_entries *chain, struct ebt_cl_stack *cl_s,
    unsigned int udc_cnt, unsigned int hooknr, char *base)
 {
 	int i, chain_nr = -1, pos = 0, nentries = chain->nentries, verdict;
-	struct ebt_entry *e = (struct ebt_entry *)chain->data;
-	struct ebt_entry_target *t;
+	const struct ebt_entry *e = (struct ebt_entry *)chain->data;
+	const struct ebt_entry_target *t;
 
 	while (pos < nentries || chain_nr != -1) {
 		/* end of udc, go back one 'recursion' step */
@@ -808,7 +810,7 @@ letscontinue:
 }
 
 /* do the parsing of the table/chains/entries/matches/watchers/targets, heh */
-static int translate_table(char *name, struct ebt_table_info *newinfo)
+static int translate_table(const char *name, struct ebt_table_info *newinfo)
 {
 	unsigned int i, j, k, udc_cnt;
 	int ret;
@@ -927,7 +929,7 @@ static int translate_table(char *name, struct ebt_table_info *newinfo)
 }
 
 /* called under write_lock */
-static void get_counters(struct ebt_counter *oldcounters,
+static void get_counters(const struct ebt_counter *oldcounters,
    struct ebt_counter *counters, unsigned int nentries)
 {
 	int i, cpu;
@@ -950,7 +952,8 @@ static void get_counters(struct ebt_counter *oldcounters,
 }
 
 /* replace the table */
-static int do_replace(struct net *net, void __user *user, unsigned int len)
+static int do_replace(struct net *net, const void __user *user,
+		      unsigned int len)
 {
 	int ret, i, countersize;
 	struct ebt_table_info *newinfo;
@@ -1230,7 +1233,8 @@ void ebt_unregister_table(struct ebt_table *table)
 }
 
 /* userspace just supplied us with counters */
-static int update_counters(struct net *net, void __user *user, unsigned int len)
+static int update_counters(struct net *net, const void __user *user,
+			   unsigned int len)
 {
 	int i, ret;
 	struct ebt_counter *tmp;
@@ -1285,8 +1289,8 @@ free_tmp:
 	return ret;
 }
 
-static inline int ebt_make_matchname(struct ebt_entry_match *m,
-   char *base, char __user *ubase)
+static inline int ebt_make_matchname(const struct ebt_entry_match *m,
+    const char *base, char __user *ubase)
 {
 	char __user *hlp = ubase + ((char *)m - base);
 	if (copy_to_user(hlp, m->u.match->name, EBT_FUNCTION_MAXNAMELEN))
@@ -1294,8 +1298,8 @@ static inline int ebt_make_matchname(struct ebt_entry_match *m,
 	return 0;
 }
 
-static inline int ebt_make_watchername(struct ebt_entry_watcher *w,
-   char *base, char __user *ubase)
+static inline int ebt_make_watchername(const struct ebt_entry_watcher *w,
+    const char *base, char __user *ubase)
 {
 	char __user *hlp = ubase + ((char *)w - base);
 	if (copy_to_user(hlp , w->u.watcher->name, EBT_FUNCTION_MAXNAMELEN))
@@ -1303,11 +1307,12 @@ static inline int ebt_make_watchername(struct ebt_entry_watcher *w,
 	return 0;
 }
 
-static inline int ebt_make_names(struct ebt_entry *e, char *base, char __user *ubase)
+static inline int
+ebt_make_names(struct ebt_entry *e, const char *base, char __user *ubase)
 {
 	int ret;
 	char __user *hlp;
-	struct ebt_entry_target *t;
+	const struct ebt_entry_target *t;
 
 	if (e->bitmask == 0)
 		return 0;
@@ -1328,10 +1333,11 @@ static inline int ebt_make_names(struct ebt_entry *e, char *base, char __user *u
 
 /* called with ebt_mutex locked */
 static int copy_everything_to_user(struct ebt_table *t, void __user *user,
-   int *len, int cmd)
+    const int *len, int cmd)
 {
 	struct ebt_replace tmp;
-	struct ebt_counter *counterstmp, *oldcounters;
+	struct ebt_counter *counterstmp;
+	const struct ebt_counter *oldcounters;
 	unsigned int entries_size, nentries;
 	char *entries;
 
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 52c96b3..95c9645 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -233,7 +233,14 @@ arpt_error(struct sk_buff *skb, const struct xt_target_param *par)
 	return NF_DROP;
 }
 
-static inline struct arpt_entry *get_entry(void *base, unsigned int offset)
+static inline const struct arpt_entry_target *
+arpt_get_target_c(const struct arpt_entry *e)
+{
+	return arpt_get_target((struct arpt_entry *)e);
+}
+
+static inline struct arpt_entry *
+get_entry(const void *base, unsigned int offset)
 {
 	return (struct arpt_entry *)(base + offset);
 }
@@ -280,7 +287,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 
 	arp = arp_hdr(skb);
 	do {
-		struct arpt_entry_target *t;
+		const struct arpt_entry_target *t;
 		int hdr_len;
 
 		if (!arp_packet_match(arp, skb->dev, indev, outdev, &e->arp)) {
@@ -292,7 +299,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 			(2 * skb->dev->addr_len);
 		ADD_COUNTER(e->counters, hdr_len, 1);
 
-		t = arpt_get_target(e);
+		t = arpt_get_target_c(e);
 
 		/* Standard target? */
 		if (!t->u.kernel.target->target) {
@@ -358,7 +365,7 @@ static inline bool unconditional(const struct arpt_arp *arp)
 /* Figures out from what hook each rule can be called: returns 0 if
  * there are loops.  Puts hook bitmask in comefrom.
  */
-static int mark_source_chains(struct xt_table_info *newinfo,
+static int mark_source_chains(const struct xt_table_info *newinfo,
 			      unsigned int valid_hooks, void *entry0)
 {
 	unsigned int hook;
@@ -379,7 +386,7 @@ static int mark_source_chains(struct xt_table_info *newinfo,
 
 		for (;;) {
 			const struct arpt_standard_target *t
-				= (void *)arpt_get_target(e);
+				= (void *)arpt_get_target_c(e);
 			int visited = e->comefrom & (1 << hook);
 
 			if (e->comefrom & (1 << NF_ARP_NUMHOOKS)) {
@@ -463,7 +470,7 @@ static int mark_source_chains(struct xt_table_info *newinfo,
 	return 1;
 }
 
-static inline int check_entry(struct arpt_entry *e, const char *name)
+static inline int check_entry(const struct arpt_entry *e, const char *name)
 {
 	const struct arpt_entry_target *t;
 
@@ -475,7 +482,7 @@ static inline int check_entry(struct arpt_entry *e, const char *name)
 	if (e->target_offset + sizeof(struct arpt_entry_target) > e->next_offset)
 		return -EINVAL;
 
-	t = arpt_get_target(e);
+	t = arpt_get_target_c(e);
 	if (e->target_offset + t->u.target_size > e->next_offset)
 		return -EINVAL;
 
@@ -540,14 +547,14 @@ out:
 	return ret;
 }
 
-static bool check_underflow(struct arpt_entry *e)
+static bool check_underflow(const struct arpt_entry *e)
 {
 	const struct arpt_entry_target *t;
 	unsigned int verdict;
 
 	if (!unconditional(&e->arp))
 		return false;
-	t = arpt_get_target(e);
+	t = arpt_get_target_c(e);
 	if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0)
 		return false;
 	verdict = ((struct arpt_standard_target *)t)->verdict;
@@ -557,8 +564,8 @@ static bool check_underflow(struct arpt_entry *e)
 
 static inline int check_entry_size_and_hooks(struct arpt_entry *e,
 					     struct xt_table_info *newinfo,
-					     unsigned char *base,
-					     unsigned char *limit,
+					     const unsigned char *base,
+					     const unsigned char *limit,
 					     const unsigned int *hook_entries,
 					     const unsigned int *underflows,
 					     unsigned int valid_hooks,
@@ -768,11 +775,11 @@ static void get_counters(const struct xt_table_info *t,
 	local_bh_enable();
 }
 
-static struct xt_counters *alloc_counters(struct xt_table *table)
+static struct xt_counters *alloc_counters(const struct xt_table *table)
 {
 	unsigned int countersize;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 
 	/* We need atomic snapshot of counters: rest doesn't change
 	 * (other than comefrom, which userspace doesn't care
@@ -790,11 +797,11 @@ static struct xt_counters *alloc_counters(struct xt_table *table)
 }
 
 static int copy_entries_to_user(unsigned int total_size,
-				struct xt_table *table,
+				const struct xt_table *table,
 				void __user *userptr)
 {
 	unsigned int off, num;
-	struct arpt_entry *e;
+	const struct arpt_entry *e;
 	struct xt_counters *counters;
 	struct xt_table_info *private = table->private;
 	int ret = 0;
@@ -814,7 +821,7 @@ static int copy_entries_to_user(unsigned int total_size,
 	/* FIXME: use iterator macros --RR */
 	/* ... then go back and fix counters and names */
 	for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){
-		struct arpt_entry_target *t;
+		const struct arpt_entry_target *t;
 
 		e = (struct arpt_entry *)(loc_cpu_entry + off);
 		if (copy_to_user(userptr + off
@@ -825,7 +832,7 @@ static int copy_entries_to_user(unsigned int total_size,
 			goto free_counters;
 		}
 
-		t = arpt_get_target(e);
+		t = arpt_get_target_c(e);
 		if (copy_to_user(userptr + off + e->target_offset
 				 + offsetof(struct arpt_entry_target,
 					    u.user.name),
@@ -860,18 +867,18 @@ static int compat_standard_to_user(void __user *dst, const void *src)
 	return copy_to_user(dst, &cv, sizeof(cv)) ? -EFAULT : 0;
 }
 
-static int compat_calc_entry(struct arpt_entry *e,
+static int compat_calc_entry(const struct arpt_entry *e,
 			     const struct xt_table_info *info,
-			     void *base, struct xt_table_info *newinfo)
+			     const void *base, struct xt_table_info *newinfo)
 {
-	struct arpt_entry_target *t;
+	const struct arpt_entry_target *t;
 	unsigned int entry_offset;
 	int off, i, ret;
 
 	off = sizeof(struct arpt_entry) - sizeof(struct compat_arpt_entry);
 	entry_offset = (void *)e - base;
 
-	t = arpt_get_target(e);
+	t = arpt_get_target_c(e);
 	off += xt_compat_target_offset(t->u.kernel.target);
 	newinfo->size -= off;
 	ret = xt_compat_add_offset(NFPROTO_ARP, entry_offset, off);
@@ -907,7 +914,8 @@ static int compat_table_info(const struct xt_table_info *info,
 }
 #endif
 
-static int get_info(struct net *net, void __user *user, int *len, int compat)
+static int get_info(struct net *net, void __user *user,
+                    const int *len, int compat)
 {
 	char name[ARPT_TABLE_MAXNAMELEN];
 	struct xt_table *t;
@@ -966,7 +974,7 @@ static int get_info(struct net *net, void __user *user, int *len, int compat)
 }
 
 static int get_entries(struct net *net, struct arpt_get_entries __user *uptr,
-		       int *len)
+		       const int *len)
 {
 	int ret;
 	struct arpt_get_entries get;
@@ -1080,7 +1088,8 @@ static int __do_replace(struct net *net, const char *name,
 	return ret;
 }
 
-static int do_replace(struct net *net, void __user *user, unsigned int len)
+static int do_replace(struct net *net, const void __user *user,
+                      unsigned int len)
 {
 	int ret;
 	struct arpt_replace tmp;
@@ -1140,8 +1149,8 @@ add_counter_to_entry(struct arpt_entry *e,
 	return 0;
 }
 
-static int do_add_counters(struct net *net, void __user *user, unsigned int len,
-			   int compat)
+static int do_add_counters(struct net *net, const void __user *user,
+			   unsigned int len, int compat)
 {
 	unsigned int i, curcpu;
 	struct xt_counters_info tmp;
@@ -1245,10 +1254,10 @@ static inline int
 check_compat_entry_size_and_hooks(struct compat_arpt_entry *e,
 				  struct xt_table_info *newinfo,
 				  unsigned int *size,
-				  unsigned char *base,
-				  unsigned char *limit,
-				  unsigned int *hook_entries,
-				  unsigned int *underflows,
+				  const unsigned char *base,
+				  const unsigned char *limit,
+				  const unsigned int *hook_entries,
+				  const unsigned int *underflows,
 				  unsigned int *i,
 				  const char *name)
 {
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 3e355fe..13574a8 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -176,7 +176,7 @@ ipt_error(struct sk_buff *skb, const struct xt_target_param *par)
 
 /* Performance critical - called for every packet */
 static inline bool
-do_match(struct ipt_entry_match *m, const struct sk_buff *skb,
+do_match(const struct ipt_entry_match *m, const struct sk_buff *skb,
 	 struct xt_match_param *par)
 {
 	par->match     = m->u.kernel.match;
@@ -191,7 +191,7 @@ do_match(struct ipt_entry_match *m, const struct sk_buff *skb,
 
 /* Performance critical */
 static inline struct ipt_entry *
-get_entry(void *base, unsigned int offset)
+get_entry(const void *base, unsigned int offset)
 {
 	return (struct ipt_entry *)(base + offset);
 }
@@ -206,6 +206,13 @@ static inline bool unconditional(const struct ipt_ip *ip)
 #undef FWINV
 }
 
+/* for const-correctness */
+static inline const struct ipt_entry_target *
+ipt_get_target_c(const struct ipt_entry *e)
+{
+	return ipt_get_target((struct ipt_entry *)e);
+}
+
 #if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
     defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
 static const char *const hooknames[] = {
@@ -240,11 +247,11 @@ static struct nf_loginfo trace_loginfo = {
 
 /* Mildly perf critical (only if packet tracing is on) */
 static inline int
-get_chainname_rulenum(struct ipt_entry *s, struct ipt_entry *e,
+get_chainname_rulenum(const struct ipt_entry *s, const struct ipt_entry *e,
 		      const char *hookname, const char **chainname,
 		      const char **comment, unsigned int *rulenum)
 {
-	struct ipt_standard_target *t = (void *)ipt_get_target(s);
+	const struct ipt_standard_target *t = (void *)ipt_get_target_c(s);
 
 	if (strcmp(t->target.u.kernel.target->name, IPT_ERROR_TARGET) == 0) {
 		/* Head of user chain: ERROR target with chainname */
@@ -270,15 +277,15 @@ get_chainname_rulenum(struct ipt_entry *s, struct ipt_entry *e,
 	return 0;
 }
 
-static void trace_packet(struct sk_buff *skb,
+static void trace_packet(const struct sk_buff *skb,
 			 unsigned int hook,
 			 const struct net_device *in,
 			 const struct net_device *out,
 			 const char *tablename,
-			 struct xt_table_info *private,
-			 struct ipt_entry *e)
+			 const struct xt_table_info *private,
+			 const struct ipt_entry *e)
 {
-	void *table_base;
+	const void *table_base;
 	const struct ipt_entry *root;
 	const char *hookname, *chainname, *comment;
 	unsigned int rulenum = 0;
@@ -322,9 +329,9 @@ ipt_do_table(struct sk_buff *skb,
 	/* Initializing verdict to NF_DROP keeps gcc happy. */
 	unsigned int verdict = NF_DROP;
 	const char *indev, *outdev;
-	void *table_base;
+	const void *table_base;
 	struct ipt_entry *e, *back;
-	struct xt_table_info *private;
+	const struct xt_table_info *private;
 	struct xt_match_param mtpar;
 	struct xt_target_param tgpar;
 
@@ -357,7 +364,7 @@ ipt_do_table(struct sk_buff *skb,
 	back = get_entry(table_base, private->underflow[hook]);
 
 	do {
-		struct ipt_entry_target *t;
+		const struct ipt_entry_target *t;
 
 		IP_NF_ASSERT(e);
 		IP_NF_ASSERT(back);
@@ -450,7 +457,7 @@ ipt_do_table(struct sk_buff *skb,
 /* Figures out from what hook each rule can be called: returns 0 if
    there are loops.  Puts hook bitmask in comefrom. */
 static int
-mark_source_chains(struct xt_table_info *newinfo,
+mark_source_chains(const struct xt_table_info *newinfo,
 		   unsigned int valid_hooks, void *entry0)
 {
 	unsigned int hook;
@@ -468,8 +475,8 @@ mark_source_chains(struct xt_table_info *newinfo,
 		e->counters.pcnt = pos;
 
 		for (;;) {
-			struct ipt_standard_target *t
-				= (void *)ipt_get_target(e);
+			const struct ipt_standard_target *t
+				= (void *)ipt_get_target_c(e);
 			int visited = e->comefrom & (1 << hook);
 
 			if (e->comefrom & (1 << NF_INET_NUMHOOKS)) {
@@ -577,9 +584,9 @@ cleanup_match(struct ipt_entry_match *m, unsigned int *i)
 }
 
 static int
-check_entry(struct ipt_entry *e, const char *name)
+check_entry(const struct ipt_entry *e, const char *name)
 {
-	struct ipt_entry_target *t;
+	const struct ipt_entry_target *t;
 
 	if (!ip_checkentry(&e->ip)) {
 		duprintf("ip_tables: ip check failed %p %s.\n", e, name);
@@ -590,7 +597,7 @@ check_entry(struct ipt_entry *e, const char *name)
 	    e->next_offset)
 		return -EINVAL;
 
-	t = ipt_get_target(e);
+	t = ipt_get_target_c(e);
 	if (e->target_offset + t->u.target_size > e->next_offset)
 		return -EINVAL;
 
@@ -715,14 +722,14 @@ find_check_entry(struct ipt_entry *e, const char *name, unsigned int size,
 	return ret;
 }
 
-static bool check_underflow(struct ipt_entry *e)
+static bool check_underflow(const struct ipt_entry *e)
 {
 	const struct ipt_entry_target *t;
 	unsigned int verdict;
 
 	if (!unconditional(&e->ip))
 		return false;
-	t = ipt_get_target(e);
+	t = ipt_get_target_c(e);
 	if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0)
 		return false;
 	verdict = ((struct ipt_standard_target *)t)->verdict;
@@ -733,8 +740,8 @@ static bool check_underflow(struct ipt_entry *e)
 static int
 check_entry_size_and_hooks(struct ipt_entry *e,
 			   struct xt_table_info *newinfo,
-			   unsigned char *base,
-			   unsigned char *limit,
+			   const unsigned char *base,
+			   const unsigned char *limit,
 			   const unsigned int *hook_entries,
 			   const unsigned int *underflows,
 			   unsigned int valid_hooks,
@@ -947,11 +954,11 @@ get_counters(const struct xt_table_info *t,
 	local_bh_enable();
 }
 
-static struct xt_counters * alloc_counters(struct xt_table *table)
+static struct xt_counters *alloc_counters(const struct xt_table *table)
 {
 	unsigned int countersize;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 
 	/* We need atomic snapshot of counters: rest doesn't change
 	   (other than comefrom, which userspace doesn't care
@@ -969,11 +976,11 @@ static struct xt_counters * alloc_counters(struct xt_table *table)
 
 static int
 copy_entries_to_user(unsigned int total_size,
-		     struct xt_table *table,
+		     const struct xt_table *table,
 		     void __user *userptr)
 {
 	unsigned int off, num;
-	struct ipt_entry *e;
+	const struct ipt_entry *e;
 	struct xt_counters *counters;
 	const struct xt_table_info *private = table->private;
 	int ret = 0;
@@ -1025,7 +1032,7 @@ copy_entries_to_user(unsigned int total_size,
 			}
 		}
 
-		t = ipt_get_target(e);
+		t = ipt_get_target_c(e);
 		if (copy_to_user(userptr + off + e->target_offset
 				 + offsetof(struct ipt_entry_target,
 					    u.user.name),
@@ -1061,24 +1068,24 @@ static int compat_standard_to_user(void __user *dst, const void *src)
 }
 
 static inline int
-compat_calc_match(struct ipt_entry_match *m, int *size)
+compat_calc_match(const struct ipt_entry_match *m, int *size)
 {
 	*size += xt_compat_match_offset(m->u.kernel.match);
 	return 0;
 }
 
-static int compat_calc_entry(struct ipt_entry *e,
+static int compat_calc_entry(const struct ipt_entry *e,
 			     const struct xt_table_info *info,
-			     void *base, struct xt_table_info *newinfo)
+			     const void *base, struct xt_table_info *newinfo)
 {
-	struct ipt_entry_target *t;
+	const struct ipt_entry_target *t;
 	unsigned int entry_offset;
 	int off, i, ret;
 
 	off = sizeof(struct ipt_entry) - sizeof(struct compat_ipt_entry);
 	entry_offset = (void *)e - base;
 	IPT_MATCH_ITERATE(e, compat_calc_match, &off);
-	t = ipt_get_target(e);
+	t = ipt_get_target_c(e);
 	off += xt_compat_target_offset(t->u.kernel.target);
 	newinfo->size -= off;
 	ret = xt_compat_add_offset(AF_INET, entry_offset, off);
@@ -1114,7 +1121,8 @@ static int compat_table_info(const struct xt_table_info *info,
 }
 #endif
 
-static int get_info(struct net *net, void __user *user, int *len, int compat)
+static int get_info(struct net *net, void __user *user,
+                    const int *len, int compat)
 {
 	char name[IPT_TABLE_MAXNAMELEN];
 	struct xt_table *t;
@@ -1174,7 +1182,8 @@ static int get_info(struct net *net, void __user *user, int *len, int compat)
 }
 
 static int
-get_entries(struct net *net, struct ipt_get_entries __user *uptr, int *len)
+get_entries(struct net *net, struct ipt_get_entries __user *uptr,
+	    const int *len)
 {
 	int ret;
 	struct ipt_get_entries get;
@@ -1284,7 +1293,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks,
 }
 
 static int
-do_replace(struct net *net, void __user *user, unsigned int len)
+do_replace(struct net *net, const void __user *user, unsigned int len)
 {
 	int ret;
 	struct ipt_replace tmp;
@@ -1345,7 +1354,8 @@ add_counter_to_entry(struct ipt_entry *e,
 }
 
 static int
-do_add_counters(struct net *net, void __user *user, unsigned int len, int compat)
+do_add_counters(struct net *net, const void __user *user,
+                unsigned int len, int compat)
 {
 	unsigned int i, curcpu;
 	struct xt_counters_info tmp;
@@ -1541,10 +1551,10 @@ static int
 check_compat_entry_size_and_hooks(struct compat_ipt_entry *e,
 				  struct xt_table_info *newinfo,
 				  unsigned int *size,
-				  unsigned char *base,
-				  unsigned char *limit,
-				  unsigned int *hook_entries,
-				  unsigned int *underflows,
+				  const unsigned char *base,
+				  const unsigned char *limit,
+				  const unsigned int *hook_entries,
+				  const unsigned int *underflows,
 				  unsigned int *i,
 				  const char *name)
 {
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 07efff1..6e7908d 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -208,7 +208,7 @@ ip6t_error(struct sk_buff *skb, const struct xt_target_param *par)
 
 /* Performance critical - called for every packet */
 static inline bool
-do_match(struct ip6t_entry_match *m, const struct sk_buff *skb,
+do_match(const struct ip6t_entry_match *m, const struct sk_buff *skb,
 	 struct xt_match_param *par)
 {
 	par->match     = m->u.kernel.match;
@@ -222,7 +222,7 @@ do_match(struct ip6t_entry_match *m, const struct sk_buff *skb,
 }
 
 static inline struct ip6t_entry *
-get_entry(void *base, unsigned int offset)
+get_entry(const void *base, unsigned int offset)
 {
 	return (struct ip6t_entry *)(base + offset);
 }
@@ -236,6 +236,12 @@ static inline bool unconditional(const struct ip6t_ip6 *ipv6)
 	return memcmp(ipv6, &uncond, sizeof(uncond)) == 0;
 }
 
+static inline const struct ip6t_entry_target *
+ip6t_get_target_c(const struct ip6t_entry *e)
+{
+	return ip6t_get_target((struct ip6t_entry *)e);
+}
+
 #if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
     defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
 /* This cries for unification! */
@@ -271,11 +277,11 @@ static struct nf_loginfo trace_loginfo = {
 
 /* Mildly perf critical (only if packet tracing is on) */
 static inline int
-get_chainname_rulenum(struct ip6t_entry *s, struct ip6t_entry *e,
+get_chainname_rulenum(const struct ip6t_entry *s, const struct ip6t_entry *e,
 		      const char *hookname, const char **chainname,
 		      const char **comment, unsigned int *rulenum)
 {
-	struct ip6t_standard_target *t = (void *)ip6t_get_target(s);
+	const struct ip6t_standard_target *t = (void *)ip6t_get_target_c(s);
 
 	if (strcmp(t->target.u.kernel.target->name, IP6T_ERROR_TARGET) == 0) {
 		/* Head of user chain: ERROR target with chainname */
@@ -301,15 +307,15 @@ get_chainname_rulenum(struct ip6t_entry *s, struct ip6t_entry *e,
 	return 0;
 }
 
-static void trace_packet(struct sk_buff *skb,
+static void trace_packet(const struct sk_buff *skb,
 			 unsigned int hook,
 			 const struct net_device *in,
 			 const struct net_device *out,
 			 const char *tablename,
-			 struct xt_table_info *private,
-			 struct ip6t_entry *e)
+			 const struct xt_table_info *private,
+			 const struct ip6t_entry *e)
 {
-	void *table_base;
+	const void *table_base;
 	const struct ip6t_entry *root;
 	const char *hookname, *chainname, *comment;
 	unsigned int rulenum = 0;
@@ -352,9 +358,9 @@ ip6t_do_table(struct sk_buff *skb,
 	/* Initializing verdict to NF_DROP keeps gcc happy. */
 	unsigned int verdict = NF_DROP;
 	const char *indev, *outdev;
-	void *table_base;
+	const void *table_base;
 	struct ip6t_entry *e, *back;
-	struct xt_table_info *private;
+	const struct xt_table_info *private;
 	struct xt_match_param mtpar;
 	struct xt_target_param tgpar;
 
@@ -385,7 +391,7 @@ ip6t_do_table(struct sk_buff *skb,
 	back = get_entry(table_base, private->underflow[hook]);
 
 	do {
-		struct ip6t_entry_target *t;
+		const struct ip6t_entry_target *t;
 
 		IP_NF_ASSERT(e);
 		IP_NF_ASSERT(back);
@@ -400,7 +406,7 @@ ip6t_do_table(struct sk_buff *skb,
 			    ntohs(ipv6_hdr(skb)->payload_len) +
 			    sizeof(struct ipv6hdr), 1);
 
-		t = ip6t_get_target(e);
+		t = ip6t_get_target_c(e);
 		IP_NF_ASSERT(t->u.kernel.target);
 
 #if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
@@ -482,7 +488,7 @@ ip6t_do_table(struct sk_buff *skb,
 /* Figures out from what hook each rule can be called: returns 0 if
    there are loops.  Puts hook bitmask in comefrom. */
 static int
-mark_source_chains(struct xt_table_info *newinfo,
+mark_source_chains(const struct xt_table_info *newinfo,
 		   unsigned int valid_hooks, void *entry0)
 {
 	unsigned int hook;
@@ -500,8 +506,8 @@ mark_source_chains(struct xt_table_info *newinfo,
 		e->counters.pcnt = pos;
 
 		for (;;) {
-			struct ip6t_standard_target *t
-				= (void *)ip6t_get_target(e);
+			const struct ip6t_standard_target *t
+				= (void *)ip6t_get_target_c(e);
 			int visited = e->comefrom & (1 << hook);
 
 			if (e->comefrom & (1 << NF_INET_NUMHOOKS)) {
@@ -609,9 +615,9 @@ cleanup_match(struct ip6t_entry_match *m, unsigned int *i)
 }
 
 static int
-check_entry(struct ip6t_entry *e, const char *name)
+check_entry(const struct ip6t_entry *e, const char *name)
 {
-	struct ip6t_entry_target *t;
+	const struct ip6t_entry_target *t;
 
 	if (!ip6_checkentry(&e->ipv6)) {
 		duprintf("ip_tables: ip check failed %p %s.\n", e, name);
@@ -622,7 +628,7 @@ check_entry(struct ip6t_entry *e, const char *name)
 	    e->next_offset)
 		return -EINVAL;
 
-	t = ip6t_get_target(e);
+	t = ip6t_get_target_c(e);
 	if (e->target_offset + t->u.target_size > e->next_offset)
 		return -EINVAL;
 
@@ -747,14 +753,14 @@ find_check_entry(struct ip6t_entry *e, const char *name, unsigned int size,
 	return ret;
 }
 
-static bool check_underflow(struct ip6t_entry *e)
+static bool check_underflow(const struct ip6t_entry *e)
 {
 	const struct ip6t_entry_target *t;
 	unsigned int verdict;
 
 	if (!unconditional(&e->ipv6))
 		return false;
-	t = ip6t_get_target(e);
+	t = ip6t_get_target_c(e);
 	if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0)
 		return false;
 	verdict = ((struct ip6t_standard_target *)t)->verdict;
@@ -765,8 +771,8 @@ static bool check_underflow(struct ip6t_entry *e)
 static int
 check_entry_size_and_hooks(struct ip6t_entry *e,
 			   struct xt_table_info *newinfo,
-			   unsigned char *base,
-			   unsigned char *limit,
+			   const unsigned char *base,
+			   const unsigned char *limit,
 			   const unsigned int *hook_entries,
 			   const unsigned int *underflows,
 			   unsigned int valid_hooks,
@@ -979,11 +985,11 @@ get_counters(const struct xt_table_info *t,
 	local_bh_enable();
 }
 
-static struct xt_counters *alloc_counters(struct xt_table *table)
+static struct xt_counters *alloc_counters(const struct xt_table *table)
 {
 	unsigned int countersize;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 
 	/* We need atomic snapshot of counters: rest doesn't change
 	   (other than comefrom, which userspace doesn't care
@@ -1001,11 +1007,11 @@ static struct xt_counters *alloc_counters(struct xt_table *table)
 
 static int
 copy_entries_to_user(unsigned int total_size,
-		     struct xt_table *table,
+		     const struct xt_table *table,
 		     void __user *userptr)
 {
 	unsigned int off, num;
-	struct ip6t_entry *e;
+	const struct ip6t_entry *e;
 	struct xt_counters *counters;
 	const struct xt_table_info *private = table->private;
 	int ret = 0;
@@ -1057,7 +1063,7 @@ copy_entries_to_user(unsigned int total_size,
 			}
 		}
 
-		t = ip6t_get_target(e);
+		t = ip6t_get_target_c(e);
 		if (copy_to_user(userptr + off + e->target_offset
 				 + offsetof(struct ip6t_entry_target,
 					    u.user.name),
@@ -1093,24 +1099,24 @@ static int compat_standard_to_user(void __user *dst, const void *src)
 }
 
 static inline int
-compat_calc_match(struct ip6t_entry_match *m, int *size)
+compat_calc_match(const struct ip6t_entry_match *m, int *size)
 {
 	*size += xt_compat_match_offset(m->u.kernel.match);
 	return 0;
 }
 
-static int compat_calc_entry(struct ip6t_entry *e,
+static int compat_calc_entry(const struct ip6t_entry *e,
 			     const struct xt_table_info *info,
-			     void *base, struct xt_table_info *newinfo)
+			     const void *base, struct xt_table_info *newinfo)
 {
-	struct ip6t_entry_target *t;
+	const struct ip6t_entry_target *t;
 	unsigned int entry_offset;
 	int off, i, ret;
 
 	off = sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry);
 	entry_offset = (void *)e - base;
 	IP6T_MATCH_ITERATE(e, compat_calc_match, &off);
-	t = ip6t_get_target(e);
+	t = ip6t_get_target_c(e);
 	off += xt_compat_target_offset(t->u.kernel.target);
 	newinfo->size -= off;
 	ret = xt_compat_add_offset(AF_INET6, entry_offset, off);
@@ -1146,7 +1152,8 @@ static int compat_table_info(const struct xt_table_info *info,
 }
 #endif
 
-static int get_info(struct net *net, void __user *user, int *len, int compat)
+static int get_info(struct net *net, void __user *user,
+                    const int *len, int compat)
 {
 	char name[IP6T_TABLE_MAXNAMELEN];
 	struct xt_table *t;
@@ -1206,7 +1213,8 @@ static int get_info(struct net *net, void __user *user, int *len, int compat)
 }
 
 static int
-get_entries(struct net *net, struct ip6t_get_entries __user *uptr, int *len)
+get_entries(struct net *net, struct ip6t_get_entries __user *uptr,
+            const int *len)
 {
 	int ret;
 	struct ip6t_get_entries get;
@@ -1317,7 +1325,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks,
 }
 
 static int
-do_replace(struct net *net, void __user *user, unsigned int len)
+do_replace(struct net *net, const void __user *user, unsigned int len)
 {
 	int ret;
 	struct ip6t_replace tmp;
@@ -1378,7 +1386,7 @@ add_counter_to_entry(struct ip6t_entry *e,
 }
 
 static int
-do_add_counters(struct net *net, void __user *user, unsigned int len,
+do_add_counters(struct net *net, const void __user *user, unsigned int len,
 		int compat)
 {
 	unsigned int i, curcpu;
@@ -1577,10 +1585,10 @@ static int
 check_compat_entry_size_and_hooks(struct compat_ip6t_entry *e,
 				  struct xt_table_info *newinfo,
 				  unsigned int *size,
-				  unsigned char *base,
-				  unsigned char *limit,
-				  unsigned int *hook_entries,
-				  unsigned int *underflows,
+				  const unsigned char *base,
+				  const unsigned char *limit,
+				  const unsigned int *hook_entries,
+				  const unsigned int *underflows,
 				  unsigned int *i,
 				  const char *name)
 {
-- 
1.6.5.2