Hi,
The current bridge multicast code uses a custom rhashtable
implementation which predates the generic rhashtable API. Patch 01
converts it to use the generic kernel rhashtable which simplifies the
code a lot and removes duplicated functionality. The convert also makes
hash_elasticity obsolete as the generic rhashtable already has such
checks and has a fixed elasticity of RHT_ELASTICITY (16 currently) so we
emit a warning whenever elasticity is set and return RHT_ELASTICITY when
read (patch 03). Patch 02 converts the multicast code to use non-bh RCU
flavor as it was mixing bh and non-bh. Since now we have the generic
rhashtable which autoshrinks we can be more liberal with the default
hash maximum so patch 04 increases it to 4096 and moves it to a define in
br_private.h.
v3: add non-rcu br_mdb_get variant and use it where we have
multicast_lock, drop special hash_max handling and just set it where
needed and use non-bh RCU consistently (patch 02, new)
v2: send the latest version of the set which handles when IGMP snooping
is not defined, changes are in patch 01
Thanks,
Nik
Nikolay Aleksandrov (4):
net: bridge: convert multicast to generic rhashtable
net: bridge: multicast: use non-bh rcu flavor
net: bridge: mark hash_elasticity as obsolete
net: bridge: increase multicast's default maximum number of entries
net/bridge/br_device.c | 10 +
net/bridge/br_mdb.c | 122 +++++------
net/bridge/br_multicast.c | 439 +++++++-------------------------------
net/bridge/br_netlink.c | 20 +-
net/bridge/br_private.h | 42 ++--
net/bridge/br_sysfs_br.c | 14 +-
6 files changed, 177 insertions(+), 470 deletions(-)
--
2.17.2
