Skip to contentSkip to conversation

reiserfs: Truncate blocks not used by a write

PATCH In review Jan Kara 2 participants 3 messages Sep 1, 2009

lore
Share & tools
lore

Jan Kara

author
#0 lore raw

PATCH 1/1 reiserfs: Truncate blocks not used by a write

It can happen that write does not use all the blocks allocated in write_begin either because of some filesystem error (like ENOSPC) or because page with data to write has been removed from memory. We truncate these blocks so that we don't have dangling blocks beyond i_size.

---

Andrew, would you merge the patch please? Thanks.

/fs/reiserfs/inode.c renamed from fs/reiserfs/inode.c +8 -3
diff -rupX /home/jack/.kerndiffexclude linux-2.6.27-SLE11_BRANCH/fs/reiserfs/inode.c linux-2.6.27-SLE11_BRANCH-1-reiserfs_corruption//fs/reiserfs/inode.c
@@ -2606,6 +2610,8 @@ static int reiserfs_write_begin(struct f
 	if (ret) {
 		unlock_page(page);
 		page_cache_release(page);
+		/* Truncate allocated blocks */
+		vmtruncate(inode, inode->i_size);
 	}
 	return ret;
 }
@@ -2698,8 +2705,7 @@ static int reiserfs_write_end(struct fil
 	 ** transaction tracking stuff when the size changes.  So, we have
 	 ** to do the i_size updates here.
 	 */
-	pos += copied;
-	if (pos > inode->i_size) {
+	if (pos + copied > inode->i_size) {
 		struct reiserfs_transaction_handle myth;
 		reiserfs_write_lock(inode->i_sb);
 		/* If the file have grown beyond the border where it
@@ -2717,7 +2723,7 @@ static int reiserfs_write_end(struct fil
 			goto journal_error;
 		}
 		reiserfs_update_inode_transaction(inode);
-		inode->i_size = pos;
+		inode->i_size = pos + copied;
 		/*
 		 * this will just nest into our transaction.  It's important
 		 * to use mark_inode_dirty so the inode gets pushed around on the
@@ -2744,6 +2750,10 @@ static int reiserfs_write_end(struct fil
       out:
 	unlock_page(page);
 	page_cache_release(page);
+
+	if (pos + len > inode->i_size)
+		vmtruncate(inode, inode->i_size);
+
 	return ret == 0 ? copied : ret;
 
       journal_error:
#1 lore raw
Quoted 4 lines · Jan Kara

It can happen that write does not use all the blocks allocated in write_begin either because of some filesystem error (like ENOSPC) or because page with data to write has been removed from memory. We truncate these blocks so that we don't have dangling blocks beyond i_size.

vmtruncate is in the process of going away for 2.6.32. Please use a filesystems-specific routine to just trim the blocks, similar to what the vmtruncate call in the generic write_begin helper is beeing replaced with in Nick's truncate patch series.

author
#2 lore raw
Quoted 10 lines · Christoph Hellwig
Quoted 4 lines · Jan Kara

It can happen that write does not use all the blocks allocated in write_begin either because of some filesystem error (like ENOSPC) or because page with data to write has been removed from memory. We truncate these blocks so that we don't have dangling blocks beyond i_size.

vmtruncate is in the process of going away for 2.6.32. Please use a filesystems-specific routine to just trim the blocks, similar to what the vmtruncate call in the generic write_begin helper is beeing replaced with in Nick's truncate patch series.

Reiserfs does not have such helper (yet), that's why I've used
vmtruncate. But OK, I'll try to convert reiserfs to the new truncate sequence and then make this fix use it...

Honza

Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR