[Cluster-devel] GFS2: Fix truncation of journaled data files

[Steven Whitehouse <swhiteho@redhat.com>]

From fa731fc4e045a801814547188a63c2cd49a4cfe6 Mon Sep 17 00:00:00 2001
From: Steven Whitehouse <swhiteho@redhat.com>
Date: Tue, 13 Nov 2012 09:50:28 +0000
Subject: GFS2: Fix truncation of journaled data files

This patch fixes an issue relating to not having enough revokes
available when truncating journaled data files. In order to ensure
that we do no run out, the truncation is broken into separate pieces
if it is large enough.

Tested using fsx on a journaled data file.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
---
 fs/gfs2/bmap.c |   52 +++++++++++++++++++++++++++++++++++++++++++++++++---
 1 files changed, 49 insertions(+), 3 deletions(-)

diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index de70e52..a68e91b 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -991,6 +991,41 @@ unlock:
 	return err;
 }
 
+/**
+ * gfs2_journaled_truncate - Wrapper for truncate_pagecache for jdata files
+ * @inode: The inode being truncated
+ * @oldsize: The original (larger) size
+ * @newsize: The new smaller size
+ *
+ * With jdata files, we have to journal a revoke for each block which is
+ * truncated. As a result, we need to split this into separate transactions
+ * if the number of pages being truncated gets too large.
+ */
+
+#define GFS2_JTRUNC_REVOKES 8192
+
+static int gfs2_journaled_truncate(struct inode *inode, u64 oldsize, u64 newsize)
+{
+	struct gfs2_sbd *sdp = GFS2_SB(inode);
+	u64 max_chunk = GFS2_JTRUNC_REVOKES * sdp->sd_vfs->s_blocksize;
+	u64 chunk;
+	int error;
+
+	while (oldsize != newsize) {
+		chunk = oldsize - newsize;
+		if (chunk > max_chunk)
+			chunk = max_chunk;
+		truncate_pagecache(inode, oldsize, oldsize - chunk);
+		oldsize -= chunk;
+		gfs2_trans_end(sdp);
+		error = gfs2_trans_begin(sdp, RES_DINODE, GFS2_JTRUNC_REVOKES);
+		if (error)
+			return error;
+	}
+
+	return 0;
+}
+
 static int trunc_start(struct inode *inode, u64 oldsize, u64 newsize)
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
@@ -1000,8 +1035,10 @@ static int trunc_start(struct inode *inode, u64 oldsize, u64 newsize)
 	int journaled = gfs2_is_jdata(ip);
 	int error;
 
-	error = gfs2_trans_begin(sdp,
-				 RES_DINODE + (journaled ? RES_JDATA : 0), 0);
+	if (journaled)
+		error = gfs2_trans_begin(sdp, RES_DINODE + RES_JDATA, GFS2_JTRUNC_REVOKES);
+	else
+		error = gfs2_trans_begin(sdp, RES_DINODE, 0);
 	if (error)
 		return error;
 
@@ -1026,7 +1063,16 @@ static int trunc_start(struct inode *inode, u64 oldsize, u64 newsize)
 	ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
 	gfs2_dinode_out(ip, dibh->b_data);
 
-	truncate_pagecache(inode, oldsize, newsize);
+	if (journaled)
+		error = gfs2_journaled_truncate(inode, oldsize, newsize);
+	else
+		truncate_pagecache(inode, oldsize, newsize);
+
+	if (error) {
+		brelse(dibh);
+		return error;
+	}
+
 out_brelse:
 	brelse(dibh);
 out:
-- 
1.7.4