SUNRPC: Don't reencode message if transmission failed with ENOBUFS
If we're running out of buffer memory when transmitting data, then
we want to just delay for a moment, and then continue transmitting
the remainder of the message.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index cbc6af9..23608eb 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -1902,6 +1902,7 @@
switch (task->tk_status) {
case -EAGAIN:
+ case -ENOBUFS:
break;
default:
dprint_status(task);
@@ -1928,7 +1929,6 @@
case -ECONNABORTED:
case -EADDRINUSE:
case -ENOTCONN:
- case -ENOBUFS:
case -EPIPE:
rpc_task_force_reencode(task);
}
@@ -2057,12 +2057,13 @@
case -ECONNABORTED:
rpc_force_rebind(clnt);
case -EADDRINUSE:
- case -ENOBUFS:
rpc_delay(task, 3*HZ);
case -EPIPE:
case -ENOTCONN:
task->tk_action = call_bind;
break;
+ case -ENOBUFS:
+ rpc_delay(task, HZ>>2);
case -EAGAIN:
task->tk_action = call_transmit;
break;