From 04b8418d2646c4663c716e6f29ff08edbb5aecd1 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Sat, 18 Oct 2008 20:28:00 -0700
Subject: [PATCH] --- yaml --- r: 115924 b: refs/heads/master c:
 5ec8b75e3a2a94860ee99b5456fe1a963c8680e5 h: refs/heads/master v: v3

---
 [refs]                | 2 +-
 trunk/fs/ext3/inode.c | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/[refs] b/[refs]
index 72042d5d7621..7213dfab0d72 100644
--- a/[refs]
+++ b/[refs]
@@ -1,2 +1,2 @@
 ---
-refs/heads/master: 6a897cf447a83c9c3fd1b85a1e525c02d6eada7d
+refs/heads/master: 5ec8b75e3a2a94860ee99b5456fe1a963c8680e5
diff --git a/trunk/fs/ext3/inode.c b/trunk/fs/ext3/inode.c
index ebfec4d0148e..f8424ad89971 100644
--- a/trunk/fs/ext3/inode.c
+++ b/trunk/fs/ext3/inode.c
@@ -1186,6 +1186,13 @@ static int ext3_write_begin(struct file *file, struct address_space *mapping,
 		ext3_journal_stop(handle);
 		unlock_page(page);
 		page_cache_release(page);
+		/*
+		 * block_write_begin may have instantiated a few blocks
+		 * outside i_size.  Trim these off again. Don't need
+		 * i_size_read because we hold i_mutex.
+		 */
+		if (pos + len > inode->i_size)
+			vmtruncate(inode, inode->i_size);
 	}
 	if (ret == -ENOSPC && ext3_should_retry_alloc(inode->i_sb, &retries))
 		goto retry;