From: Steven Whitehouse <swhiteho@redhat.com>
Subject: [RHEL5.1] [GFS2] Fix bz 231380: GFS2 will hang if you run iozone on one node and do a du -h on another node
Date: Thu, 05 Apr 2007 11:13:59 +0100
Bugzilla: 231380
Message-Id: <1175768039.1636.238.camel@quoit.chygwyn.com>
Changelog: [GFS2] resolve deadlock when writing and accessing a file


Hi,

This patch (from Josef Whiter) fixes the problem that we have been
having with deadlocks during unlocking a glock when there is a queued,
pending demote request. Its already in the upstream kernel and has been
verified to fix the bug.


Steve.


I'm not entirely sure if this is the best fix for this problem, so I'm open for
suggetions.  If we are writing a file, and in the middle of writing the file
another node attempts to get a shared lock on that file (by doing a du for
example) the process doing the writing will hang waiting on lock_page.  The
reason for this is because when we have waiters on a exclusive glock, we will go
through and flush out all dirty pages associated with that inode and release the
lock.  The problem is that when we flush the dirty pages, we could hit a page
that we have locked durring the generic_file_buffered_write part of this
operation.  This patch unlocks the page before we go to dequeue the lock and
locks it immediatly afterwards, since generic_file_buffered_write needs the page
locked when the commit_write is completed.  This patch resolves the problem,
however if somebody sees a better way to do this please don't hesistate to yell.
Thank you,

Josef


Signed-off-by: Josef Whiter <jwhiter@redhat.com>


--- gfs2-2.6-nmw/fs/gfs2/ops_address.c.josef	2007-03-08 15:04:10.000000000 -0500
+++ gfs2-2.6-nmw/fs/gfs2/ops_address.c	2007-03-12 17:23:24.000000000 -0400
@@ -507,7 +507,9 @@
 		gfs2_quota_unlock(ip);
 		gfs2_alloc_put(ip);
 	}
+	unlock_page(page);
 	gfs2_glock_dq_m(1, &ip->i_gh);
+	lock_page(page);
 	gfs2_holder_uninit(&ip->i_gh);
 	return 0;
 
@@ -520,7 +522,9 @@
 		gfs2_quota_unlock(ip);
 		gfs2_alloc_put(ip);
 	}
+	unlock_page(page);
 	gfs2_glock_dq_m(1, &ip->i_gh);
+	lock_page(page);
 	gfs2_holder_uninit(&ip->i_gh);
 fail_nounlock:
 	ClearPageUptodate(page);