public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* Has anyone run the Connectathon Testsuite recently?
@ 2001-10-22  6:24 H . J . Lu
  2001-10-22  7:54 ` Alan Cox
  2001-10-22  8:28 ` [NFS] " Trond Myklebust
  0 siblings, 2 replies; 3+ messages in thread
From: H . J . Lu @ 2001-10-22  6:24 UTC (permalink / raw)
  To: nfs; +Cc: linux kernel, alan

I checked out kernel 2.4.9-6 from RedHat 7.1 updates. It failed the
Connectathon Testsuite against the Linux and none-Linux server. I
believe both NFS server and client are broken in 2.4.9-6. See

http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=54868

Now the question is how bad the current Linus/AC kernels are?


H.J.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Has anyone run the Connectathon Testsuite recently?
  2001-10-22  6:24 Has anyone run the Connectathon Testsuite recently? H . J . Lu
@ 2001-10-22  7:54 ` Alan Cox
  2001-10-22  8:28 ` [NFS] " Trond Myklebust
  1 sibling, 0 replies; 3+ messages in thread
From: Alan Cox @ 2001-10-22  7:54 UTC (permalink / raw)
  To: H . J . Lu; +Cc: nfs, linux kernel, alan

> Connectathon Testsuite against the Linux and none-Linux server. I
> believe both NFS server and client are broken in 2.4.9-6. See
> 
> http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=54868
> 
> Now the question is how bad the current Linus/AC kernels are?

Hopefully pretty close since there isn't much NFS related difference between
vanilla and -ac currently. The -ac client has the NFS inode structure in its 
own slab being the major change

Alan

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [NFS] Has anyone run the Connectathon Testsuite recently?
  2001-10-22  6:24 Has anyone run the Connectathon Testsuite recently? H . J . Lu
  2001-10-22  7:54 ` Alan Cox
@ 2001-10-22  8:28 ` Trond Myklebust
  1 sibling, 0 replies; 3+ messages in thread
From: Trond Myklebust @ 2001-10-22  8:28 UTC (permalink / raw)
  To: H . J . Lu; +Cc: nfs, linux kernel, alan

>>>>> " " == hjl  <H> writes:

     > I checked out kernel 2.4.9-6 from RedHat 7.1 updates. It failed
     > the Connectathon Testsuite against the Linux and none-Linux
     > server. I believe both NFS server and client are broken in
     > 2.4.9-6. See

     > http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=54868

     > Now the question is how bad the current Linus/AC kernels are?


They are not affected. The RedHat kernel seems to have the (known) bug
in which the grace period isn't reset. My fault for introducing it in
the 2.4.9-ac series...

I haven't gotten round to syncing up the AC kernel to the full reclaim
code that's in Linus' kernel, but Alan has already applied the
following patch (as of 2.4.10-ac12).

Cheers,
  Trond

--- linux-2.4.9-6/fs/lockd/svc.c.orig	Thu Oct 18 15:00:46 2001
+++ linux-2.4.9-6/fs/lockd/svc.c	Mon Oct 22 10:25:21 2001
@@ -122,6 +122,15 @@
 			if (nlmsvc_ops) {
 				nlmsvc_ops->detach();
 				grace_period_expire = nlmsvc_grace_period + jiffies;
+#ifdef RPC_DEBUG
+				nlmsvc_grace_period = 10 * HZ;
+#else
+				if (nlm_grace_period)
+					nlmsvc_grace_period = ((nlm_grace_period + nlm_timeout - 1)
+								/ nlm_timeout) * nlm_timeout * HZ;
+				else
+					nlmsvc_grace_period = 5 * nlm_timeout * HZ;
+#endif
 			}
 		}
 
@@ -133,8 +142,10 @@
 		 */
 		if (!grace_period_expire) {
 			timeout = nlmsvc_retry_blocked();
-		} else if (time_before(grace_period_expire, jiffies))
+		} else if (time_before(grace_period_expire, jiffies)) {
 			grace_period_expire = 0;
+			nlmsvc_grace_period = 0;
+		}
 
 		/*
 		 * Find a socket with data available and call its


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2001-10-22  8:28 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-10-22  6:24 Has anyone run the Connectathon Testsuite recently? H . J . Lu
2001-10-22  7:54 ` Alan Cox
2001-10-22  8:28 ` [NFS] " Trond Myklebust

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox