Subject : nfsd deadlock, 2.6.36-rc3
Submitter : Tim Gardner <email@example.com>
Date : 2010-09-01 15:39
Message-ID : 4C7E73CB.firstname.lastname@example.org
References : http://marc.info/?l=linux-kernel&m=128335681711984&w=2
This entry is being used for tracking a regression from 2.6.35. Please don't
close it until the problem is fixed in the mainline.
To me that looks just like the usual deadlock scenario that is known to occur when the nfs server enters direct page reclaim on a loopback-mounted nfs client.
Reassigning to Bruce.
Trond is correct, see followup in mailing list discussion.