[EXT4 set 9][PATCH 5/5]Extent micro cleanups

[EXT4 set 9][PATCH 5/5]Extent micro cleanups

[Mingming Cao]

From: Dmitry Monakhov <dmonakhov@sw.ru>
Subject: ext4: extent macros cleanup

- Replace math equation to it's macro equivalent
- make ext4_ext_grow_indepth() indexes/leaf correct

Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Acked-by: Alex Tomas <alex@clusterfs.com>
Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
---
 fs/ext4/extents.c |   11 +++++++----
 1 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 12fe3d7..1fd00ac 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -375,7 +375,7 @@ ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int bloc
 	ext_debug("binsearch for %d(idx):  ", block);

 	l = EXT_FIRST_INDEX(eh) + 1;
-	r = EXT_FIRST_INDEX(eh) + le16_to_cpu(eh->eh_entries) - 1;
+	r = EXT_LAST_INDEX(eh);
 	while (l <= r) {
 		m = l + (r - l) / 2;
 		if (block < le32_to_cpu(m->ei_block))
@@ -440,7 +440,7 @@ ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block)
 	ext_debug("binsearch for %d:  ", block);

 	l = EXT_FIRST_EXTENT(eh) + 1;
-	r = EXT_FIRST_EXTENT(eh) + le16_to_cpu(eh->eh_entries) - 1;
+	r = EXT_LAST_EXTENT(eh);

 	while (l <= r) {
 		m = l + (r - l) / 2;
@@ -922,8 +922,11 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode,
 	curp->p_hdr->eh_max = cpu_to_le16(ext4_ext_space_root_idx(inode));
 	curp->p_hdr->eh_entries = cpu_to_le16(1);
 	curp->p_idx = EXT_FIRST_INDEX(curp->p_hdr);
-	/* FIXME: it works, but actually path[0] can be index */
-	curp->p_idx->ei_block = EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block;
+	
+	if (path[0].p_hdr->eh_depth)
+	  curp->p_idx->ei_block = EXT_FIRST_INDEX(path[0].p_hdr)->ei_block;
+	else
+	  curp->p_idx->ei_block = EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block;
 	ext4_idx_store_pblock(curp->p_idx, newblock);

 	neh = ext_inode_hdr(inode);

Re: [EXT4 set 9][PATCH 5/5]Extent micro cleanups

[Andrew Morton]

On Sun, 01 Jul 2007 03:38:59 -0400 Mingming Cao <cmm@us.ibm.com> wrote:

> From: Dmitry Monakhov <dmonakhov@sw.ru>
> Subject: ext4: extent macros cleanup
> 
> - Replace math equation to it's macro equivalent

s/it's/its/;)

> - make ext4_ext_grow_indepth() indexes/leaf correct

hm, what was wrong with it?

> Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
> Acked-by: Alex Tomas <alex@clusterfs.com>
> Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
> ---
>  fs/ext4/extents.c |   11 +++++++----
>  1 files changed, 7 insertions(+), 4 deletions(-)
> 
> diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
> index 12fe3d7..1fd00ac 100644
> --- a/fs/ext4/extents.c
> +++ b/fs/ext4/extents.c
> @@ -375,7 +375,7 @@ ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int bloc
>  	ext_debug("binsearch for %d(idx):  ", block);
> 
>  	l = EXT_FIRST_INDEX(eh) + 1;
> -	r = EXT_FIRST_INDEX(eh) + le16_to_cpu(eh->eh_entries) - 1;
> +	r = EXT_LAST_INDEX(eh);
>  	while (l <= r) {
>  		m = l + (r - l) / 2;
>  		if (block < le32_to_cpu(m->ei_block))
> @@ -440,7 +440,7 @@ ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block)
>  	ext_debug("binsearch for %d:  ", block);
> 
>  	l = EXT_FIRST_EXTENT(eh) + 1;
> -	r = EXT_FIRST_EXTENT(eh) + le16_to_cpu(eh->eh_entries) - 1;
> +	r = EXT_LAST_EXTENT(eh);
> 
>  	while (l <= r) {
>  		m = l + (r - l) / 2;
> @@ -922,8 +922,11 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode,
>  	curp->p_hdr->eh_max = cpu_to_le16(ext4_ext_space_root_idx(inode));
>  	curp->p_hdr->eh_entries = cpu_to_le16(1);
>  	curp->p_idx = EXT_FIRST_INDEX(curp->p_hdr);
> -	/* FIXME: it works, but actually path[0] can be index */
> -	curp->p_idx->ei_block = EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block;
> +	
> +	if (path[0].p_hdr->eh_depth)
> +	  curp->p_idx->ei_block = EXT_FIRST_INDEX(path[0].p_hdr)->ei_block;
> +	else
> +	  curp->p_idx->ei_block = EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block;

whitespace bustage there.

Re: [EXT4 set 9][PATCH 5/5]Extent micro cleanups

[Mingming Cao]

On Tue, 2007-07-10 at 23:20 -0700, Andrew Morton wrote:
> On Sun, 01 Jul 2007 03:38:59 -0400 Mingming Cao <cmm@us.ibm.com> wrote:
> 
> > From: Dmitry Monakhov <dmonakhov@sw.ru>
> > Subject: ext4: extent macros cleanup
> > 
> > - Replace math equation to it's macro equivalent
> 
> s/it's/its/;)
Okay.

> 
> > - make ext4_ext_grow_indepth() indexes/leaf correct
> 
> hm, what was wrong with it?
> 
Looking at the code, ext4_ext_ext_grow_indepth() implements tree growing
procedure. It allocates a new index block, moves the top-level data of
the tree(root or leaf blocks in i_data) into the new block, initializes
the new root, creating index that points to the just created index block

The original top-level data (in i_data) could be extent tree root (index
block) or extents (leaf block). The current code (without the patch)
treats the top-level data always be the leaf block, which is incorrect.


assumes when the tree is growing the extent structure pass in is always
> > @@ -922,8 +922,11 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode,
> >  	curp->p_hdr->eh_max = cpu_to_le16(ext4_ext_space_root_idx(inode));
> >  	curp->p_hdr->eh_entries = cpu_to_le16(1);
> >  	curp->p_idx = EXT_FIRST_INDEX(curp->p_hdr);
> > -	/* FIXME: it works, but actually path[0] can be index */
> > -	curp->p_idx->ei_block = EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block;
> > +	
> > +	if (path[0].p_hdr->eh_depth)
> > +	  curp->p_idx->ei_block = EXT_FIRST_INDEX(path[0].p_hdr)->ei_block;
> > +	else
> > +	  curp->p_idx->ei_block = EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block;
> 
> whitespace bustage there.
> 
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html