Neil,
Here is the same patch for 2.4.* hopefuly you can get this in the kernel as
well as I have not received any reply from Marcelo.
Juan
----- Forwarded by Juan Gomez/Almaden/IBM on 10/16/02 10:00 AM -----
Juan Gomez
To: [email protected]
10/15/02 09:10 AM cc: [email protected]
From: Juan Gomez/Almaden/IBM@IBMUS
Subject: kNFS(lockd) patch for linux-2.4.19
Marcelo,
Have you had a chance to include this?
Juan
----- Forwarded by Juan Gomez/Almaden/IBM on 10/15/02 09:11 AM -----
Juan Gomez
<[email protected]> To: [email protected]
Sent by: cc: [email protected], [email protected]
linux-kernel-owner@vger Subject: kNFS(lockd) patch for linux-2.4.19
.kernel.org
10/08/02 01:59 PM
Marcelo,
Would you please consider the attached patch for inclusion in 2.4..*?
The patch solves a faulty delay observed by the first client that access
lockd just after the grace period.
Juan
diff -ru linux-2.4.19/fs/lockd/svc.c
linux-2.4.19-plus-delay-patch/fs/lockd/svc.c
--- linux-2.4.19/fs/lockd/svc.c Sun Oct 21 10:32:33 2001
+++ linux-2.4.19-plus-delay-patch/fs/lockd/svc.c Tue Oct 8
13:19:40 2002
@@ -144,8 +144,7 @@
*/
if (!nlmsvc_grace_period) {
timeout = nlmsvc_retry_blocked();
- } else if (time_before(grace_period_expire,
jiffies))
- clear_grace_period();
+ }
/*
* Find a socket with data available and call its
@@ -163,6 +162,22 @@
dprintk("lockd: request from %08x\n",
(unsigned)ntohl(rqstp->rq_addr.sin_addr.s_addr));
+ /*
+ * We need to do the clear/grace period here and not
before
+ * svc_recv() because svc_recv() may sleep longer than the
+ * grace period and the first request may be falsely
processed
+ * as if the server was in the grace period when it was
not
+ * causing unnecessary delays for the first request
received.
+ * Juan C. Gomez [email protected]
+ */
+
+ if (nlmsvc_grace_period
+ &&
+ time_before(grace_period_expire, jiffies)) {
+ clear_grace_period();
+ }
+
+
/*
* Look up the NFS client handle. The handle is
needed for