| Message ID | 1426952290-18962-2-git-send-email-kaber@trash.net |
|---|---|
| State | Awaiting Upstream, archived |
| Delegated to: | David Miller |
| Headers | show |
On Sat, Mar 21, 2015 at 03:38:09PM +0000, Patrick McHardy wrote: > nftables sets will be converted to use so called setextensions, moving > the key to a non-fixed position. To hash it, the obj_hashfn must be used, > however it so far doesn't receive the length parameter. > > Pass the key length to obj_hashfn() and convert existing users. > > Signed-off-by: Patrick McHardy <kaber@trash.net> I presume you've got the offset embedded inside the object. Can you show us what your actual code looks like? I'm wondering if you could instead provide rhashtable with a fixed offset key with the object offset embedded in the key that then lets you go back up to figure out the object? Cheers,
On 03/21/15 at 03:38pm, Patrick McHardy wrote: > nftables sets will be converted to use so called setextensions, moving > the key to a non-fixed position. To hash it, the obj_hashfn must be used, > however it so far doesn't receive the length parameter. > > Pass the key length to obj_hashfn() and convert existing users. > > Signed-off-by: Patrick McHardy <kaber@trash.net> Acked-by: Thomas Graf <tgraf@suug.ch> -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/include/linux/rhashtable.h b/include/linux/rhashtable.h index 89d1022..d94dd53 100644 --- a/include/linux/rhashtable.h +++ b/include/linux/rhashtable.h @@ -87,7 +87,7 @@ struct rhashtable_compare_arg { }; typedef u32 (*rht_hashfn_t)(const void *data, u32 len, u32 seed); -typedef u32 (*rht_obj_hashfn_t)(const void *data, u32 seed); +typedef u32 (*rht_obj_hashfn_t)(const void *data, u32 len, u32 seed); typedef int (*rht_obj_cmpfn_t)(struct rhashtable_compare_arg *arg, const void *obj); @@ -211,7 +211,9 @@ static inline unsigned int rht_head_hashfn( const char *ptr = rht_obj(ht, he); return likely(params.obj_hashfn) ? - rht_bucket_index(tbl, params.obj_hashfn(ptr, tbl->hash_rnd)) : + rht_bucket_index(tbl, params.obj_hashfn(ptr, params.key_len ?: + ht->p.key_len, + tbl->hash_rnd)) : rht_key_hashfn(ht, tbl, ptr + params.key_offset, params); } diff --git a/lib/rhashtable.c b/lib/rhashtable.c index 83cfedd..c897671 100644 --- a/lib/rhashtable.c +++ b/lib/rhashtable.c @@ -559,7 +559,7 @@ static size_t rounded_hashtable_size(const struct rhashtable_params *params) * struct rhash_head node; * }; * - * u32 my_hash_fn(const void *data, u32 seed) + * u32 my_hash_fn(const void *data, u32 len, u32 seed) * { * struct test_obj *obj = data; * diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 6517921..f15b651 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -3127,7 +3127,7 @@ static struct pernet_operations __net_initdata netlink_net_ops = { .exit = netlink_net_exit, }; -static inline u32 netlink_hash(const void *data, u32 seed) +static inline u32 netlink_hash(const void *data, u32 len, u32 seed) { const struct netlink_sock *nlk = data; struct netlink_compare_arg arg;
nftables sets will be converted to use so called setextensions, moving the key to a non-fixed position. To hash it, the obj_hashfn must be used, however it so far doesn't receive the length parameter. Pass the key length to obj_hashfn() and convert existing users. Signed-off-by: Patrick McHardy <kaber@trash.net> --- include/linux/rhashtable.h | 6 ++++-- lib/rhashtable.c | 2 +- net/netlink/af_netlink.c | 2 +- 3 files changed, 6 insertions(+), 4 deletions(-)