[GFS2] Fix lock ordering bug in page fault path
Mmapped files were able to trigger a lock ordering bug. Private
maps do not need to take the glock so early on. Shared maps do
unfortunately, however we can get around that by adding a flag
into the flags for the struct gfs2_file. This only works because
we are taking an exclusive lock at this point, so we know that
nobody else can be racing with us.
Fixes Red Hat bugzilla: #201196
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c
index 7aabc03..bbd44a4 100644
--- a/fs/gfs2/recovery.c
+++ b/fs/gfs2/recovery.c
@@ -153,8 +153,7 @@
if (lh.lh_header.mh_magic != GFS2_MAGIC ||
lh.lh_header.mh_type != GFS2_METATYPE_LH ||
- lh.lh_blkno != blk ||
- lh.lh_hash != hash)
+ lh.lh_blkno != blk || lh.lh_hash != hash)
return 1;
*head = lh;
@@ -482,11 +481,9 @@
/* Acquire a shared hold on the transaction lock */
- error = gfs2_glock_nq_init(sdp->sd_trans_gl,
- LM_ST_SHARED,
+ error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED,
LM_FLAG_NOEXP | LM_FLAG_PRIORITY |
- GL_NOCANCEL | GL_NOCACHE,
- &t_gh);
+ GL_NOCANCEL | GL_NOCACHE, &t_gh);
if (error)
goto fail_gunlock_ji;