diff options
author | Andy Grover <andy.grover@oracle.com> | 2010-03-11 13:50:04 +0000 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2010-03-16 21:16:59 -0700 |
commit | b98ba52f96e7cdb4dbe2b06bced83d95d94c9d02 (patch) | |
tree | 72e22772c6326ff554bd030bda35abcb1526249e /net/rds/rds.h | |
parent | 550a8002e4340eaf3bc333e33b59427e9c20272d (diff) |
RDS: only put sockets that have seen congestion on the poll_waitq
rds_poll_waitq's listeners will be awoken if we receive a congestion
notification. Bad performance may result because *all* polled sockets
contend for this single lock. However, it should not be necessary to
wake pollers when a congestion update arrives if they have never
experienced congestion, and not putting these on the waitq will
hopefully greatly reduce contention.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/rds/rds.h')
-rw-r--r-- | net/rds/rds.h | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/net/rds/rds.h b/net/rds/rds.h index 85d6f897ecc..4bec6e2ed49 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -388,6 +388,8 @@ struct rds_sock { /* flag indicating we were congested or not */ int rs_congested; + /* seen congestion (ENOBUFS) when sending? */ + int rs_seen_congestion; /* rs_lock protects all these adjacent members before the newline */ spinlock_t rs_lock; |