From a1b08e75dff3dc18a88444803753e667bb1d126e Mon Sep 17 00:00:00 2001 From: Tao Ma Date: Thu, 27 Aug 2009 14:46:56 +0800 Subject: ocfs2: invalidate dentry if its dentry_lock isn't initialized. In commit a5a0a630922a2f6a774b6dac19f70cb5abd86bb0, when ocfs2_attch_dentry_lock fails, we call an extra iput and reset dentry->d_fsdata to NULL. This resolve a bug, but it isn't completed and the dentry is still there. When we want to use it again, ocfs2_dentry_revalidate doesn't catch it and return true. That make future ocfs2_dentry_lock panic out. One bug is http://oss.oracle.com/bugzilla/show_bug.cgi?id=1162. The resolution is to add a check for dentry->d_fsdata in revalidate process and return false if dentry->d_fsdata is NULL, so that a new ocfs2_lookup will be called again. Signed-off-by: Tao Ma Signed-off-by: Joel Becker --- fs/ocfs2/dcache.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'fs') diff --git a/fs/ocfs2/dcache.c b/fs/ocfs2/dcache.c index 2f28b7de2c8d..b4957c7d9fe2 100644 --- a/fs/ocfs2/dcache.c +++ b/fs/ocfs2/dcache.c @@ -85,6 +85,17 @@ static int ocfs2_dentry_revalidate(struct dentry *dentry, goto bail; } + /* + * If the last lookup failed to create dentry lock, let us + * redo it. + */ + if (!dentry->d_fsdata) { + mlog(0, "Inode %llu doesn't have dentry lock, " + "returning false\n", + (unsigned long long)OCFS2_I(inode)->ip_blkno); + goto bail; + } + ret = 1; bail: -- cgit v1.2.3 From 8379e7c46cc48f51197dd663fc6676f47f2a1e71 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Fri, 4 Sep 2009 11:12:01 -0700 Subject: ocfs2: ocfs2_write_begin_nolock() should handle len=0 Bug introduced by mainline commit e7432675f8ca868a4af365759a8d4c3779a3d922 The bug causes ocfs2_write_begin_nolock() to oops when len=0. Signed-off-by: Sunil Mushran Cc: stable@kernel.org Signed-off-by: Joel Becker --- fs/ocfs2/aops.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index b401654011a2..8a1e61545f41 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -1747,8 +1747,8 @@ int ocfs2_write_begin_nolock(struct address_space *mapping, * we know zeros will only be needed in the first and/or last cluster. */ if (clusters_to_alloc || extents_to_split || - wc->w_desc[0].c_needs_zero || - wc->w_desc[wc->w_clen - 1].c_needs_zero) + (wc->w_clen && (wc->w_desc[0].c_needs_zero || + wc->w_desc[wc->w_clen - 1].c_needs_zero))) cluster_of_pages = 1; else cluster_of_pages = 0; -- cgit v1.2.3