[GFS2] Fix releasepage bug (fixes direct i/o writes)

This patch fixes three main bugs. Firstly the direct i/o get_block
was returning the wrong return code in certain cases. Secondly, the
GFS2's releasepage function was not dealing with cases when clean,
ordered buffers were found still queued on a transaction (which can
happen depending on the ordering of journal flushes). Thirdly, the
journaling code itself needed altering to take account of the
after effects of removing the clean ordered buffers from the transactions
before a journal flush.

The releasepage bug did also show up under "normal" buffered i/o
as well, so its not just a fix for direct i/o. In fact its not
normally used in the direct i/o path at all, except when flushing
existing buffers after performing a direct i/o write, but that was
the code path that led us to spot this.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

2 files changed, 40 insertions, 4 deletions

diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c
index 08de8b7fb316..0ec38b399097 100644
--- a/fs/gfs2/lops.c
+++ b/fs/gfs2/lops.c
@@ -592,6 +592,17 @@ static void databuf_lo_before_commit(struct gfs2_sbd *sdp)
 				gfs2_log_lock(sdp);
 				if (n++ > num)
 					break;
+			} else if (!bd1->bd_bh) {
+				total_dbuf--;
+				sdp->sd_log_num_databuf--;
+				list_del_init(&bd1->bd_le.le_list);
+				if (bd1 == bd2) {
+					bd2 = NULL;
+					bd2 = list_prepare_entry(bd2,
+						&sdp->sd_log_le_databuf,
+						bd_le.le_list);
+                                }
+				kmem_cache_free(gfs2_bufdata_cachep, bd1);
 			}
 		}
 		gfs2_log_unlock(sdp);
diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c
index 48720421c796..0de7a9523633 100644
--- a/fs/gfs2/ops_address.c
+++ b/fs/gfs2/ops_address.c
@@ -118,6 +118,25 @@ static int get_block_noalloc(struct inode *inode, sector_t lblock,
 	return error;
 }
 
+static int get_block_direct(struct inode *inode, sector_t lblock,
+			    struct buffer_head *bh_result, int create)
+{
+	int new = 0;
+	u64 dblock;
+	int error, boundary;
+
+	error = gfs2_block_map(inode, lblock, &new, &dblock, &boundary);
+	if (error)
+		return error;
+
+	if (dblock) {
+		map_bh(bh_result, inode->i_sb, dblock);
+		if (boundary)
+			set_buffer_boundary(bh_result);
+	}
+
+	return 0;
+}
 /**
  * gfs2_writepage - Write complete page
  * @page: Page to write
@@ -661,7 +680,7 @@ static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb,
 	rv = blockdev_direct_IO_own_locking(rw, iocb, inode,
 					    inode->i_sb->s_bdev,
 					    iov, offset, nr_segs,
-					    gfs2_get_block, NULL);
+					    get_block_direct, NULL);
 out:
 	gfs2_glock_dq_m(1, &gh);
 	gfs2_holder_uninit(&gh);
@@ -724,7 +743,7 @@ static unsigned limit = 0;
 }
 
 /**
- * gfs2_aspace_releasepage - free the metadata associated with a page
+ * gfs2_releasepage - free the metadata associated with a page
  * @page: the page that's being released
  * @gfp_mask: passed from Linux VFS, ignored by us
  *
@@ -761,16 +780,22 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask)
 		}
 
 		gfs2_assert_warn(sdp, !buffer_pinned(bh));
+		gfs2_assert_warn(sdp, !buffer_dirty(bh));
 
+		gfs2_log_lock(sdp);
 		bd = bh->b_private;
 		if (bd) {
 			gfs2_assert_warn(sdp, bd->bd_bh == bh);
 			gfs2_assert_warn(sdp, list_empty(&bd->bd_list_tr));
-			gfs2_assert_warn(sdp, list_empty(&bd->bd_le.le_list));
 			gfs2_assert_warn(sdp, !bd->bd_ail);
-			kmem_cache_free(gfs2_bufdata_cachep, bd);
+			bd->bd_bh = NULL;
+			if (!list_empty(&bd->bd_le.le_list))
+				bd = NULL;
 			bh->b_private = NULL;
 		}
+		gfs2_log_unlock(sdp);
+		if (bd)
+			kmem_cache_free(gfs2_bufdata_cachep, bd);
 
 		bh = bh->b_this_page;
 	} while (bh != head);