A-Team_Digital_Solutions/android_kernel_motorola_sm6375
Watch
2
0
Fork
You've already forked android_kernel_motorola_sm6375
 0
mirror of https://github.com/BobTheBlinker/android_kernel_motorola_sm6375.git synced 2026-10-28 18:51:56 -04:00
Code Issues Projects Releases Packages Wiki Activity

rhashtable: Per bucket locks & deferred expansion/shrinking

Browse source 
Introduces an array of spinlocks to protect bucket mutations. The number
of spinlocks per CPU is configurable and selected based on the hash of
the bucket. This allows for parallel insertions and removals of entries
which do not share a lock.

The patch also defers expansion and shrinking to a worker queue which
allows insertion and removal from atomic context. Insertions and
deletions may occur in parallel to it and are only held up briefly
while the particular bucket is linked or unzipped.

Mutations of the bucket table pointer is protected by a new mutex, read
access is RCU protected.

In the event of an expansion or shrinking, the new bucket table allocated
is exposed as a so called future table as soon as the resize process
starts.  Lookups, deletions, and insertions will briefly use both tables.
The future table becomes the main table after an RCU grace period and
initial linking of the old to the new table was performed. Optimization
of the chains to make use of the new number of buckets follows only the
new table is in use.

The side effect of this is that during that RCU grace period, a bucket
traversal using any rht_for_each() variant on the main table will not see
any insertions performed during the RCU grace period which would at that
point land in the future table. The lookup will see them as it searches
both tables if needed.

Having multiple insertions and removals occur in parallel requires nelems
to become an atomic counter.

Signed-off-by: Thomas Graf <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
This commit is contained in:
Thomas Graf 2015-01-02 23:00:20 +01:00 • committed by David S. Miller
commit 97defe1ecf
4 changed files with 384 additions and 153 deletions
Download patch file Download diff file Expand all files Collapse all files

15
net/netlink/af_netlink.c
View file

@ -114,15 +114,6 @@ static atomic_t nl_table_users = ATOMIC_INIT(0);
DEFINE_MUTEX(nl_sk_hash_lock);
EXPORT_SYMBOL_GPL(nl_sk_hash_lock);
#ifdef CONFIG_PROVE_LOCKING
static int lockdep_nl_sk_hash_is_held(void *parent)
{
if (debug_locks)
return lockdep_is_held(&nl_sk_hash_lock) || lockdep_is_held(&nl_table_lock);
return 1;
}
#endif
static ATOMIC_NOTIFIER_HEAD(netlink_chain);
static DEFINE_SPINLOCK(netlink_tap_lock);
@ -1063,7 +1054,8 @@ static int netlink_insert(struct sock *sk, struct net *net, u32 portid)
goto err;
err = -ENOMEM;
if (BITS_PER_LONG > 32 && unlikely(table->hash.nelems >= UINT_MAX))
if (BITS_PER_LONG > 32 &&
unlikely(atomic_read(&table->hash.nelems) >= UINT_MAX))
goto err;
nlk_sk(sk)->portid = portid;
@ -3122,9 +3114,6 @@ static int __init netlink_proto_init(void)
.max_shift = 16, /* 64K */
.grow_decision = rht_grow_above_75,
.shrink_decision = rht_shrink_below_30,
#ifdef CONFIG_PROVE_LOCKING
.mutex_is_held = lockdep_nl_sk_hash_is_held,
#endif
};
if (err != 0)