Skip to contentSkip to conversation

ext4: move ext4_ind_migrate() into migrate.c

PATCH In review Lukas Czerner 4 participants 10 messages Apr 9, 2013 – Apr 11, 2013 ext4 file system

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/2 ext4: move ext4_ind_migrate() into migrate.c

Move ext4_ind_migrate() into migrate.c file since it makes much more sense and ext4_ext_migrate() is there as well.

Also fix tiny style problem - add spaces around "=" in "i=0".

---
v2: added to the series

 fs/ext4/ext4.h    |    2 +-
 fs/ext4/extents.c |   59 ----------------------------------------------------
 fs/ext4/migrate.c |   60 ++++++++++++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 60 insertions(+), 61 deletions(-)
fs/ext4/ext4.h +1 -1
index a0637e5..bb56920 100644
@@ -2131,6 +2131,7 @@ extern long ext4_compat_ioctl(struct file *, unsigned int, unsigned long);
 
 /* migrate.c */
 extern int ext4_ext_migrate(struct inode *);
+extern int ext4_ind_migrate(struct inode *inode);
 
 /* namei.c */
 extern int ext4_dirent_csum_verify(struct inode *inode,
@@ -2620,7 +2621,6 @@ extern int ext4_find_delalloc_range(struct inode *inode,
 extern int ext4_find_delalloc_cluster(struct inode *inode, ext4_lblk_t lblk);
 extern int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			__u64 start, __u64 len);
-extern int ext4_ind_migrate(struct inode *inode);
 
 
 /* move_extent.c */
fs/ext4/extents.c +0 -59

62 lines not on this page · raw message

fs/ext4/migrate.c +59 -1

69 lines not on this page · raw message

Signature · 1 line
1.7.7.6
#1 lore raw

PATCH v2 2/2 ext4: Do not convert to indirect with bigalloc enabled

With bigalloc feature enabled we do not support indirect addressing at all so we have to prevent extent addressing to indirect addressing conversion in this case. The problem has been introduced with the commit "ext4: support simple conversion of extent-mapped inodes to use i_blocks"

---
v2: added to the series

 fs/ext4/migrate.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)
fs/ext4/migrate.c +4 -0
index 3464604..f13a289 100644
@@ -625,6 +625,10 @@ int ext4_ind_migrate(struct inode *inode)
 	    (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))
 		return -EINVAL;
 
+	if (EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
+				       EXT4_FEATURE_RO_COMPAT_BIGALLOC))
+		return -EOPNOTSUPP;
+
 	down_write(&EXT4_I(inode)->i_data_sem);
 	ret = ext4_ext_check_inode(inode);
 	if (ret)
Signature · 1 line
1.7.7.6
Reviewed-by on v2 2/2
#2 lore raw

Re: [PATCH 2/2 v2] ext4: Do not convert to indirect with bigalloc enabled

Quoted 6 lines · Lukas Czerner

With bigalloc feature enabled we do not support indirect addressing at all so we have to prevent extent addressing to indirect addressing conversion in this case. The problem has been introduced with the commit "ext4: support simple conversion of extent-mapped inodes to use i_blocks"

Regards,
- Zheng

Quoted 24 lines

---
v2: added to the series

 fs/ext4/migrate.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)
fs/ext4/migrate.c +4 -0
index 3464604..f13a289 100644
@@ -625,6 +625,10 @@ int ext4_ind_migrate(struct inode *inode)
 	    (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))
 		return -EINVAL;
 
+	if (EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
+				       EXT4_FEATURE_RO_COMPAT_BIGALLOC))
+		return -EOPNOTSUPP;
+
 	down_write(&EXT4_I(inode)->i_data_sem);
 	ret = ext4_ext_check_inode(inode);
 	if (ret)
Signature · 1 line
1.7.7.6
maintainer
#3 lore raw

Re: [PATCH 2/2 v2] ext4: Do not convert to indirect with bigalloc enabled

In reply to Lukas Czerner
Quoted 6 lines · Lukas Czerner

With bigalloc feature enabled we do not support indirect addressing at all so we have to prevent extent addressing to indirect addressing conversion in this case. The problem has been introduced with the commit "ext4: support simple conversion of extent-mapped inodes to use i_blocks"

Thanks, this version is now in the ext4 patch queue.

- Ted

Reviewed-by on v2 1/2
#4 lore raw
In reply to Lukas Czerner
Quoted 6 lines · Lukas Czerner

Move ext4_ind_migrate() into migrate.c file since it makes much more sense and ext4_ext_migrate() is there as well.

Also fix tiny style problem - add spaces around "=" in "i=0".

Regards,
- Zheng

Quoted 173 lines

---
v2: added to the series

 fs/ext4/ext4.h    |    2 +-
 fs/ext4/extents.c |   59 ----------------------------------------------------
 fs/ext4/migrate.c |   60 ++++++++++++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 60 insertions(+), 61 deletions(-)
fs/ext4/ext4.h +1 -1
index a0637e5..bb56920 100644
@@ -2131,6 +2131,7 @@ extern long ext4_compat_ioctl(struct file *, unsigned int, unsigned long);
 
 /* migrate.c */
 extern int ext4_ext_migrate(struct inode *);
+extern int ext4_ind_migrate(struct inode *inode);
 
 /* namei.c */
 extern int ext4_dirent_csum_verify(struct inode *inode,
@@ -2620,7 +2621,6 @@ extern int ext4_find_delalloc_range(struct inode *inode,
 extern int ext4_find_delalloc_cluster(struct inode *inode, ext4_lblk_t lblk);
 extern int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			__u64 start, __u64 len);
-extern int ext4_ind_migrate(struct inode *inode);
 
 
 /* move_extent.c */
fs/ext4/extents.c +0 -59
index 6c5a70a..3645357 100644
@@ -4715,62 +4715,3 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 
 	return error;
 }
-
-/*
- * Migrate a simple extent-based inode to use the i_blocks[] array
- */
-int ext4_ind_migrate(struct inode *inode)
-{
-	struct ext4_extent_header	*eh;
-	struct ext4_super_block		*es = EXT4_SB(inode->i_sb)->s_es;
-	struct ext4_inode_info		*ei = EXT4_I(inode);
-	struct ext4_extent		*ex;
-	unsigned int			i, len;
-	ext4_fsblk_t			blk;
-	handle_t			*handle;
-	int				ret;
-
-	if (!EXT4_HAS_INCOMPAT_FEATURE(inode->i_sb,
-				       EXT4_FEATURE_INCOMPAT_EXTENTS) ||
-	    (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))
-		return -EINVAL;
-
-	down_write(&EXT4_I(inode)->i_data_sem);
-	ret = ext4_ext_check_inode(inode);
-	if (ret)
-		goto errout;
-
-	eh = ext_inode_hdr(inode);
-	ex  = EXT_FIRST_EXTENT(eh);
-	if (ext4_blocks_count(es) > EXT4_MAX_BLOCK_FILE_PHYS ||
-	    eh->eh_depth != 0 || eh->eh_entries > 1) {
-		ret = -EOPNOTSUPP;
-		goto errout;
-	}
-	if (eh->eh_entries == 0)
-		blk = len = 0;
-	else {
-		len = le16_to_cpu(ex->ee_len);
-		blk = ext4_ext_pblock(ex);
-		if (len > EXT4_NDIR_BLOCKS) {
-			ret = -EOPNOTSUPP;
-			goto errout;
-		}
-	}
-
-	handle = ext4_journal_start(inode, EXT4_HT_MIGRATE, 1);
-	if (IS_ERR(handle)) {
-		ret = PTR_ERR(handle);
-		goto errout;
-	}
-
-	ext4_clear_inode_flag(inode, EXT4_INODE_EXTENTS);
-	memset(ei->i_data, 0, sizeof(ei->i_data));
-	for (i=0; i < len; i++)
-		ei->i_data[i] = cpu_to_le32(blk++);
-	ext4_mark_inode_dirty(handle, inode);
-	ext4_journal_stop(handle);
-errout:
-	up_write(&EXT4_I(inode)->i_data_sem);
-	return ret;
-}
fs/ext4/migrate.c +59 -1
index 480acf4..3464604 100644
@@ -426,7 +426,6 @@ static int free_ext_block(handle_t *handle, struct inode *inode)
 			return retval;
 	}
 	return retval;
-
 }
 
 int ext4_ext_migrate(struct inode *inode)
@@ -606,3 +605,62 @@ out:
 
 	return retval;
 }
+
+/*
+ * Migrate a simple extent-based inode to use the i_blocks[] array
+ */
+int ext4_ind_migrate(struct inode *inode)
+{
+	struct ext4_extent_header	*eh;
+	struct ext4_super_block		*es = EXT4_SB(inode->i_sb)->s_es;
+	struct ext4_inode_info		*ei = EXT4_I(inode);
+	struct ext4_extent		*ex;
+	unsigned int			i, len;
+	ext4_fsblk_t			blk;
+	handle_t			*handle;
+	int				ret;
+
+	if (!EXT4_HAS_INCOMPAT_FEATURE(inode->i_sb,
+				       EXT4_FEATURE_INCOMPAT_EXTENTS) ||
+	    (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))
+		return -EINVAL;
+
+	down_write(&EXT4_I(inode)->i_data_sem);
+	ret = ext4_ext_check_inode(inode);
+	if (ret)
+		goto errout;
+
+	eh = ext_inode_hdr(inode);
+	ex  = EXT_FIRST_EXTENT(eh);
+	if (ext4_blocks_count(es) > EXT4_MAX_BLOCK_FILE_PHYS ||
+	    eh->eh_depth != 0 || eh->eh_entries > 1) {
+		ret = -EOPNOTSUPP;
+		goto errout;
+	}
+	if (eh->eh_entries == 0)
+		blk = len = 0;
+	else {
+		len = le16_to_cpu(ex->ee_len);
+		blk = ext4_ext_pblock(ex);
+		if (len > EXT4_NDIR_BLOCKS) {
+			ret = -EOPNOTSUPP;
+			goto errout;
+		}
+	}
+
+	handle = ext4_journal_start(inode, EXT4_HT_MIGRATE, 1);
+	if (IS_ERR(handle)) {
+		ret = PTR_ERR(handle);
+		goto errout;
+	}
+
+	ext4_clear_inode_flag(inode, EXT4_INODE_EXTENTS);
+	memset(ei->i_data, 0, sizeof(ei->i_data));
+	for (i = 0; i < len; i++)
+		ei->i_data[i] = cpu_to_le32(blk++);
+	ext4_mark_inode_dirty(handle, inode);
+	ext4_journal_stop(handle);
+errout:
+	up_write(&EXT4_I(inode)->i_data_sem);
+	return ret;
+}
Signature · 1 line
1.7.7.6
maintainer
#5 lore raw
In reply to Lukas Czerner
Quoted 6 lines · Lukas Czerner

Move ext4_ind_migrate() into migrate.c file since it makes much more sense and ext4_ext_migrate() is there as well.

Also fix tiny style problem - add spaces around "=" in "i=0".

I had to regenerate this patch since ext4_ind_migrate has been changed by other patches in the ext4 patch queue, but it's applied now.

- Ted

#6 lore raw
Quoted 10 lines · Theodore Ts'o
Quoted 6 lines · Lukas Czerner

Move ext4_ind_migrate() into migrate.c file since it makes much more sense and ext4_ext_migrate() is there as well.

Also fix tiny style problem - add spaces around "=" in "i=0".

I had to regenerate this patch since ext4_ind_migrate has been changed by other patches in the ext4 patch queue, but it's applied now.

You have placed it in to wrong position, Obviously check should be performed before journal_start()

Quoted 7 lines

- Ted
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#7 lore raw
In reply to Theodore Ts'o
Quoted 10 lines · Theodore Ts'o

Date: Wed, 10 Apr 2013 23:58:26 -0400
From: Theodore Ts'o <[email protected]>
To: Lukas Czerner <[email protected]>
Cc: [email protected], [email protected]
Subject: Re: [PATCH 1/2 v2] ext4: move ext4_ind_migrate() into migrate.c

Quoted 3 lines · Lukas Czerner

Move ext4_ind_migrate() into migrate.c file since it makes much more sense and ext4_ext_migrate() is there as well.

Quoted 6 lines
Quoted 3 lines

Also fix tiny style problem - add spaces around "=" in "i=0".

I had to regenerate this patch since ext4_ind_migrate has been changed by other patches in the ext4 patch queue, but it's applied now.

Thanks, but Dmitry is right, the check should be before the journal_start() call.

I am not sure how much work will it be for you, but would it be possible to resend patch to the list when you change it ? Just to make sure we do not get unnecessary problems.

Thanks!
-Lukas

Quoted 3 lines

- Ted

maintainer
#8 lore raw
Quoted 2 lines · Lukáš Czerner

Thanks, but Dmitry is right, the check should be before the journal_start() call.

Whoops! I'm not sure how I screwed that up. Thanks for catching that, Dmitry!

Quoted 3 lines

I am not sure how much work will it be for you, but would it be possible to resend patch to the list when you change it ? Just to make sure we do not get unnecessary problems.

Yes, I'll try to do that in the future. Here's the revised patch....

- Ted

commit 3b296c66fe6f9a6ed44b0b83f4f2ae87d8874af6

Author: Lukas Czerner <[email protected]>
Date:   Thu Apr 11 10:54:46 2013 -0400

ext4: do not convert to indirect with bigalloc enabled

With bigalloc feature enabled we do not support indirect addressing at all so we have to prevent extent addressing to indirect addressing conversion in this case. The problem has been introduced with the commit "ext4: support simple conversion of extent-mapped inodes to use i_blocks"

fs/ext4/migrate.c +4 -0
index d129a4d..49e8bdf 100644
@@ -625,6 +625,10 @@ int ext4_ind_migrate(struct inode *inode)
 	    (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))
 		return -EINVAL;
 
+	if (EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
+				       EXT4_FEATURE_RO_COMPAT_BIGALLOC))
+		return -EOPNOTSUPP;
+
 	handle = ext4_journal_start(inode, EXT4_HT_MIGRATE, 1);
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);

--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#9 lore raw
Quoted 10 lines · Theodore Ts'o

Date: Thu, 11 Apr 2013 10:56:20 -0400
From: Theodore Ts'o <[email protected]>
To: Lukáš Czerner <[email protected]>
Cc: [email protected], [email protected]
Subject: Re: [PATCH 1/2 v2] ext4: move ext4_ind_migrate() into migrate.c

Quoted 2 lines · Lukáš Czerner

Thanks, but Dmitry is right, the check should be before the journal_start() call.

Quoted 9 lines

Whoops! I'm not sure how I screwed that up. Thanks for catching that, Dmitry!

Quoted 3 lines

I am not sure how much work will it be for you, but would it be possible to resend patch to the list when you change it ? Just to make sure we do not get unnecessary problems.

Yes, I'll try to do that in the future. Here's the revised patch....

Thanks! Now it looks ok.

-Lukas

Quoted 32 lines

- Ted

commit 3b296c66fe6f9a6ed44b0b83f4f2ae87d8874af6

Author: Lukas Czerner <[email protected]>
Date:   Thu Apr 11 10:54:46 2013 -0400

ext4: do not convert to indirect with bigalloc enabled

With bigalloc feature enabled we do not support indirect addressing at all so we have to prevent extent addressing to indirect addressing conversion in this case. The problem has been introduced with the commit "ext4: support simple conversion of extent-mapped inodes to use i_blocks"

fs/ext4/migrate.c +4 -0
index d129a4d..49e8bdf 100644
@@ -625,6 +625,10 @@ int ext4_ind_migrate(struct inode *inode)
 	    (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))
 		return -EINVAL;
 
+	if (EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
+				       EXT4_FEATURE_RO_COMPAT_BIGALLOC))
+		return -EOPNOTSUPP;
+
 	handle = ext4_journal_start(inode, EXT4_HT_MIGRATE, 1);
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);