diff options
| author | Chuck Lever <chuck.lever@oracle.com> | 2017-08-23 17:05:58 -0400 |
|---|---|---|
| committer | Trond Myklebust <trond.myklebust@primarydata.com> | 2017-09-05 18:27:07 -0400 |
| commit | 9590d083c1bb1419b7992609d1a0a3e3517d3893 (patch) | |
| tree | ed285e1635c2232d66d432835fc13890ef209718 /net/sunrpc/xprtrdma/verbs.c | |
| parent | f9773b22a27a4234f436c9570afd62d905e00a13 (diff) | |
xprtrdma: Use xprt_pin_rqst in rpcrdma_reply_handler
Adopt the use of xprt_pin_rqst to eliminate contention between
Call-side users of rb_lock and the use of rb_lock in
rpcrdma_reply_handler.
This replaces the mechanism introduced in 431af645cf66 ("xprtrdma:
Fix client lock-up after application signal fires").
Use recv_lock to quickly find the completing rqst, pin it, then
drop the lock. At that point invalidation and pull-up of the Reply
XDR can be done. Both are often expensive operations.
Finally, take recv_lock again to signal completion to the RPC
layer. It also protects adjustment of "cwnd".
This greatly reduces the amount of time a lock is held by the
reply handler. Comparing lock_stat results shows a marked decrease
in contention on rb_lock and recv_lock.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
[trond.myklebust@primarydata.com: Remove call to rpcrdma_buffer_put() from
the "out_norqst:" path in rpcrdma_reply_handler.]
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Diffstat (limited to 'net/sunrpc/xprtrdma/verbs.c')
| -rw-r--r-- | net/sunrpc/xprtrdma/verbs.c | 1 |
1 files changed, 0 insertions, 1 deletions
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index c78fb27c20ed..11a1fbf7e59e 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -1001,7 +1001,6 @@ rpcrdma_buffer_create(struct rpcrdma_xprt *r_xprt) spin_lock_init(&buf->rb_recovery_lock); INIT_LIST_HEAD(&buf->rb_mws); INIT_LIST_HEAD(&buf->rb_all); - INIT_LIST_HEAD(&buf->rb_pending); INIT_LIST_HEAD(&buf->rb_stale_mrs); INIT_DELAYED_WORK(&buf->rb_refresh_worker, rpcrdma_mr_refresh_worker); |