netfilter: ipset: let destroy callbacks adjust ext mem size
Subsystems (2)
Florian Westphal
maintainerPATCH 1/1 netfilter: ipset: let destroy callbacks adjust ext mem size
For bitmap this change makes no difference, because destructors are called synchronously.
List type however calls them via call_rcu() so accounting decrement can happen after list_set_flush() set ext_size to 0.
'set->elements = 0' can be removed for the same reason in the list type case, it calls 'set->elements--' for each element.
---
sashiko: list_set_uadd() may call list_set_replace() and then erronously increments set->elements, causing a counter drift. This bug will be resolved in a different patch.
net/netfilter/ipset/ip_set_bitmap_gen.h | 2 +- net/netfilter/ipset/ip_set_list_set.c | 3 +-- 2 files changed, 2 insertions(+), 3 deletions(-)
net/netfilter/ipset/ip_set_bitmap_gen.h +1 -1
@@ -77,7 +77,7 @@ mtype_flush(struct ip_set *set)mtype_ext_cleanup (set);bitmap_zero (map->members, map->elements); set->elements =0 ; -atomic64_set (&set->ext_size,0 ); +DEBUG_NET_WARN_ON_ONCE (atomic64_read (&set->ext_size) >0 ); }/* Calculate the actual memory size of the set data */
net/netfilter/ipset/ip_set_list_set.c +1 -2
@@ -423,8 +423,7 @@ list_set_flush(struct ip_set *set)list_for_each_entry_safe (e, n, &map->members, list)list_set_del (set, e); - set->elements =0 ; -atomic64_set (&set->ext_size,0 ); +DEBUG_NET_WARN_ON_ONCE (set->elements >0 ); } static void
Signature · 1 line
2.54.0
Florian Westphal
maintainerQuoted 8 lines · Florian Westphal
For bitmap this change makes no difference, because destructors are called synchronously.
List type however calls them viacall_rcu()so accounting decrement can happen afterlist_set_flush()setext_sizeto 0.'
set->elements= 0' can be removed for the same reason in the list type case, it calls 'set->elements--' for each element.
Quoted 7 lines
Fixes 9e41f26a505c ("netfilter: ipset: Count non-static extension memory for userspace")Suggested-by Jozsef Kadlecsik <[email protected]>Signed-off-by Florian Westphal <[email protected]>---
sashiko:list_set_uadd()may calllist_set_replace()and then erronously incrementsset->elements, causing a counter drift. This bug will be resolved in a different patch.
Seems sashiko ignored this and reported this bug again. The report can thus be discarded.
Quoted 6 lines · Florian Westphal
For bitmap this change makes no difference, because destructors are called synchronously.
List type however calls them via
call_rcu()so accounting decrement can happen afterlist_set_flush()setext_sizeto 0.
Quoted 6 lines
'
set->elements= 0' can be removed for the same reason in the list type case, it calls 'set->elements--' for each element.Fixes 9e41f26a505c ("netfilter: ipset: Count non-static extension memory for userspace")Suggested-by Jozsef Kadlecsik <[email protected]>Signed-off-by Florian Westphal <[email protected]>
Best regards,
Jozsef
Quoted 40 lines
---
sashiko:list_set_uadd()may calllist_set_replace()and then erronously incrementsset->elements, causing a counter drift. This bug will be resolved in a different patch.net/netfilter/ipset/ip_set_bitmap_gen.h | 2 +- net/netfilter/ipset/ip_set_list_set.c | 3 +-- 2 files changed, 2 insertions(+), 3 deletions(-)net/netfilter/ipset/ip_set_bitmap_gen.h +1
-1@@ -77,7 +77,7 @@ mtype_flush(struct ip_set *set)mtype_ext_cleanup (set);bitmap_zero (map->members, map->elements); set->elements =0 ; -atomic64_set (&set->ext_size,0 ); +DEBUG_NET_WARN_ON_ONCE (atomic64_read (&set->ext_size) >0 ); }/* Calculate the actual memory size of the set data */ net/netfilter/ipset/ip_set_list_set.c +1
-2@@ -423,8 +423,7 @@ list_set_flush(struct ip_set *set)list_for_each_entry_safe (e, n, &map->members, list)list_set_del (set, e); - set->elements =0 ; -atomic64_set (&set->ext_size,0 ); +DEBUG_NET_WARN_ON_ONCE (set->elements >0 ); } static voidSignature · 1 line
2.54.0
Signature · 3 lines
E-mail : [email protected], [email protected], [email protected] Address: Wigner Research Centre for Physics H-1525 Budapest 114, POB. 49, Hungary