2006-10-11 04:21:03 -04:00
|
|
|
/*
|
|
|
|
* Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com
|
|
|
|
* Written by Alex Tomas <alex@clusterfs.com>
|
|
|
|
*
|
|
|
|
* Architecture independence:
|
|
|
|
* Copyright (c) 2005, Bull S.A.
|
|
|
|
* Written by Pierre Peiffer <pierre.peiffer@bull.net>
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or modify
|
|
|
|
* it under the terms of the GNU General Public License version 2 as
|
|
|
|
* published by the Free Software Foundation.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public Licens
|
|
|
|
* along with this program; if not, write to the Free Software
|
|
|
|
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Extents support for EXT4
|
|
|
|
*
|
|
|
|
* TODO:
|
|
|
|
* - ext4*_error() should be used in some situations
|
|
|
|
* - analyze all BUG()/BUG_ON(), use -EIO where appropriate
|
|
|
|
* - smart tree reduction
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/fs.h>
|
|
|
|
#include <linux/time.h>
|
2007-10-16 18:38:25 -04:00
|
|
|
#include <linux/jbd2.h>
|
2006-10-11 04:21:03 -04:00
|
|
|
#include <linux/highuid.h>
|
|
|
|
#include <linux/pagemap.h>
|
|
|
|
#include <linux/quotaops.h>
|
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/slab.h>
|
2007-07-17 21:42:41 -04:00
|
|
|
#include <linux/falloc.h>
|
2006-10-11 04:21:03 -04:00
|
|
|
#include <asm/uaccess.h>
|
2008-04-29 18:13:32 -04:00
|
|
|
#include "ext4_jbd2.h"
|
|
|
|
#include "ext4_extents.h"
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/*
|
|
|
|
* ext_pblock:
|
|
|
|
* combine low and high parts of physical block number into ext4_fsblk_t
|
|
|
|
*/
|
2006-12-06 23:41:36 -05:00
|
|
|
static ext4_fsblk_t ext_pblock(struct ext4_extent *ex)
|
2006-10-11 04:21:05 -04:00
|
|
|
{
|
|
|
|
ext4_fsblk_t block;
|
|
|
|
|
2007-10-16 18:38:25 -04:00
|
|
|
block = le32_to_cpu(ex->ee_start_lo);
|
2006-10-11 04:21:13 -04:00
|
|
|
block |= ((ext4_fsblk_t) le16_to_cpu(ex->ee_start_hi) << 31) << 1;
|
2006-10-11 04:21:05 -04:00
|
|
|
return block;
|
|
|
|
}
|
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/*
|
|
|
|
* idx_pblock:
|
|
|
|
* combine low and high parts of a leaf physical block number into ext4_fsblk_t
|
|
|
|
*/
|
2008-01-28 23:58:26 -05:00
|
|
|
ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix)
|
2006-10-11 04:21:05 -04:00
|
|
|
{
|
|
|
|
ext4_fsblk_t block;
|
|
|
|
|
2007-10-16 18:38:25 -04:00
|
|
|
block = le32_to_cpu(ix->ei_leaf_lo);
|
2006-10-11 04:21:13 -04:00
|
|
|
block |= ((ext4_fsblk_t) le16_to_cpu(ix->ei_leaf_hi) << 31) << 1;
|
2006-10-11 04:21:05 -04:00
|
|
|
return block;
|
|
|
|
}
|
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/*
|
|
|
|
* ext4_ext_store_pblock:
|
|
|
|
* stores a large physical block number into an extent struct,
|
|
|
|
* breaking it into parts
|
|
|
|
*/
|
2008-01-28 23:58:26 -05:00
|
|
|
void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb)
|
2006-10-11 04:21:05 -04:00
|
|
|
{
|
2007-10-16 18:38:25 -04:00
|
|
|
ex->ee_start_lo = cpu_to_le32((unsigned long) (pb & 0xffffffff));
|
2006-10-11 04:21:13 -04:00
|
|
|
ex->ee_start_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff);
|
2006-10-11 04:21:05 -04:00
|
|
|
}
|
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/*
|
|
|
|
* ext4_idx_store_pblock:
|
|
|
|
* stores a large physical block number into an index struct,
|
|
|
|
* breaking it into parts
|
|
|
|
*/
|
2006-12-06 23:41:36 -05:00
|
|
|
static void ext4_idx_store_pblock(struct ext4_extent_idx *ix, ext4_fsblk_t pb)
|
2006-10-11 04:21:05 -04:00
|
|
|
{
|
2007-10-16 18:38:25 -04:00
|
|
|
ix->ei_leaf_lo = cpu_to_le32((unsigned long) (pb & 0xffffffff));
|
2006-10-11 04:21:13 -04:00
|
|
|
ix->ei_leaf_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff);
|
2006-10-11 04:21:05 -04:00
|
|
|
}
|
|
|
|
|
2008-07-11 19:27:31 -04:00
|
|
|
static int ext4_ext_journal_restart(handle_t *handle, int needed)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
int err;
|
|
|
|
|
|
|
|
if (handle->h_buffer_credits > needed)
|
2008-07-11 19:27:31 -04:00
|
|
|
return 0;
|
|
|
|
err = ext4_journal_extend(handle, needed);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
return ext4_journal_restart(handle, needed);
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* could return:
|
|
|
|
* - EROFS
|
|
|
|
* - ENOMEM
|
|
|
|
*/
|
|
|
|
static int ext4_ext_get_access(handle_t *handle, struct inode *inode,
|
|
|
|
struct ext4_ext_path *path)
|
|
|
|
{
|
|
|
|
if (path->p_bh) {
|
|
|
|
/* path points to block */
|
|
|
|
return ext4_journal_get_write_access(handle, path->p_bh);
|
|
|
|
}
|
|
|
|
/* path points to leaf/index in inode body */
|
|
|
|
/* we use in-core data, no need to protect them */
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* could return:
|
|
|
|
* - EROFS
|
|
|
|
* - ENOMEM
|
|
|
|
* - EIO
|
|
|
|
*/
|
|
|
|
static int ext4_ext_dirty(handle_t *handle, struct inode *inode,
|
|
|
|
struct ext4_ext_path *path)
|
|
|
|
{
|
|
|
|
int err;
|
|
|
|
if (path->p_bh) {
|
|
|
|
/* path points to block */
|
|
|
|
err = ext4_journal_dirty_metadata(handle, path->p_bh);
|
|
|
|
} else {
|
|
|
|
/* path points to leaf/index in inode body */
|
|
|
|
err = ext4_mark_inode_dirty(handle, inode);
|
|
|
|
}
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2006-10-11 04:21:05 -04:00
|
|
|
static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode,
|
2006-10-11 04:21:03 -04:00
|
|
|
struct ext4_ext_path *path,
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t block)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
struct ext4_inode_info *ei = EXT4_I(inode);
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_fsblk_t bg_start;
|
2008-02-15 13:43:07 -05:00
|
|
|
ext4_fsblk_t last_block;
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_grpblk_t colour;
|
2006-10-11 04:21:03 -04:00
|
|
|
int depth;
|
|
|
|
|
|
|
|
if (path) {
|
|
|
|
struct ext4_extent *ex;
|
|
|
|
depth = path->p_depth;
|
|
|
|
|
|
|
|
/* try to predict block placement */
|
2006-12-06 23:41:33 -05:00
|
|
|
ex = path[depth].p_ext;
|
|
|
|
if (ex)
|
2006-10-11 04:21:05 -04:00
|
|
|
return ext_pblock(ex)+(block-le32_to_cpu(ex->ee_block));
|
2006-10-11 04:21:03 -04:00
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/* it looks like index is empty;
|
|
|
|
* try to find starting block from index itself */
|
2006-10-11 04:21:03 -04:00
|
|
|
if (path[depth].p_bh)
|
|
|
|
return path[depth].p_bh->b_blocknr;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* OK. use inode's group */
|
|
|
|
bg_start = (ei->i_block_group * EXT4_BLOCKS_PER_GROUP(inode->i_sb)) +
|
|
|
|
le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_first_data_block);
|
2008-02-15 13:43:07 -05:00
|
|
|
last_block = ext4_blocks_count(EXT4_SB(inode->i_sb)->s_es) - 1;
|
|
|
|
|
|
|
|
if (bg_start + EXT4_BLOCKS_PER_GROUP(inode->i_sb) <= last_block)
|
|
|
|
colour = (current->pid % 16) *
|
2006-10-11 04:21:03 -04:00
|
|
|
(EXT4_BLOCKS_PER_GROUP(inode->i_sb) / 16);
|
2008-02-15 13:43:07 -05:00
|
|
|
else
|
|
|
|
colour = (current->pid % 16) * ((last_block - bg_start) / 16);
|
2006-10-11 04:21:03 -04:00
|
|
|
return bg_start + colour + block;
|
|
|
|
}
|
|
|
|
|
2008-07-11 19:27:31 -04:00
|
|
|
/*
|
|
|
|
* Allocation for a meta data block
|
|
|
|
*/
|
2006-10-11 04:21:05 -04:00
|
|
|
static ext4_fsblk_t
|
2008-07-11 19:27:31 -04:00
|
|
|
ext4_ext_new_meta_block(handle_t *handle, struct inode *inode,
|
2006-10-11 04:21:03 -04:00
|
|
|
struct ext4_ext_path *path,
|
|
|
|
struct ext4_extent *ex, int *err)
|
|
|
|
{
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_fsblk_t goal, newblock;
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
goal = ext4_ext_find_goal(inode, path, le32_to_cpu(ex->ee_block));
|
2008-07-11 19:27:31 -04:00
|
|
|
newblock = ext4_new_meta_block(handle, inode, goal, err);
|
2006-10-11 04:21:03 -04:00
|
|
|
return newblock;
|
|
|
|
}
|
|
|
|
|
2006-12-06 23:41:36 -05:00
|
|
|
static int ext4_ext_space_block(struct inode *inode)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
int size;
|
|
|
|
|
|
|
|
size = (inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header))
|
|
|
|
/ sizeof(struct ext4_extent);
|
2007-02-17 13:20:16 -05:00
|
|
|
#ifdef AGGRESSIVE_TEST
|
2006-10-11 04:21:03 -04:00
|
|
|
if (size > 6)
|
|
|
|
size = 6;
|
|
|
|
#endif
|
|
|
|
return size;
|
|
|
|
}
|
|
|
|
|
2006-12-06 23:41:36 -05:00
|
|
|
static int ext4_ext_space_block_idx(struct inode *inode)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
int size;
|
|
|
|
|
|
|
|
size = (inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header))
|
|
|
|
/ sizeof(struct ext4_extent_idx);
|
2007-02-17 13:20:16 -05:00
|
|
|
#ifdef AGGRESSIVE_TEST
|
2006-10-11 04:21:03 -04:00
|
|
|
if (size > 5)
|
|
|
|
size = 5;
|
|
|
|
#endif
|
|
|
|
return size;
|
|
|
|
}
|
|
|
|
|
2006-12-06 23:41:36 -05:00
|
|
|
static int ext4_ext_space_root(struct inode *inode)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
int size;
|
|
|
|
|
|
|
|
size = sizeof(EXT4_I(inode)->i_data);
|
|
|
|
size -= sizeof(struct ext4_extent_header);
|
|
|
|
size /= sizeof(struct ext4_extent);
|
2007-02-17 13:20:16 -05:00
|
|
|
#ifdef AGGRESSIVE_TEST
|
2006-10-11 04:21:03 -04:00
|
|
|
if (size > 3)
|
|
|
|
size = 3;
|
|
|
|
#endif
|
|
|
|
return size;
|
|
|
|
}
|
|
|
|
|
2006-12-06 23:41:36 -05:00
|
|
|
static int ext4_ext_space_root_idx(struct inode *inode)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
int size;
|
|
|
|
|
|
|
|
size = sizeof(EXT4_I(inode)->i_data);
|
|
|
|
size -= sizeof(struct ext4_extent_header);
|
|
|
|
size /= sizeof(struct ext4_extent_idx);
|
2007-02-17 13:20:16 -05:00
|
|
|
#ifdef AGGRESSIVE_TEST
|
2006-10-11 04:21:03 -04:00
|
|
|
if (size > 4)
|
|
|
|
size = 4;
|
|
|
|
#endif
|
|
|
|
return size;
|
|
|
|
}
|
|
|
|
|
2008-07-14 17:52:37 -04:00
|
|
|
/*
|
|
|
|
* Calculate the number of metadata blocks needed
|
|
|
|
* to allocate @blocks
|
|
|
|
* Worse case is one block per extent
|
|
|
|
*/
|
|
|
|
int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks)
|
|
|
|
{
|
|
|
|
int lcap, icap, rcap, leafs, idxs, num;
|
|
|
|
int newextents = blocks;
|
|
|
|
|
|
|
|
rcap = ext4_ext_space_root_idx(inode);
|
|
|
|
lcap = ext4_ext_space_block(inode);
|
|
|
|
icap = ext4_ext_space_block_idx(inode);
|
|
|
|
|
|
|
|
/* number of new leaf blocks needed */
|
|
|
|
num = leafs = (newextents + lcap - 1) / lcap;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Worse case, we need separate index block(s)
|
|
|
|
* to link all new leaf blocks
|
|
|
|
*/
|
|
|
|
idxs = (leafs + icap - 1) / icap;
|
|
|
|
do {
|
|
|
|
num += idxs;
|
|
|
|
idxs = (idxs + icap - 1) / icap;
|
|
|
|
} while (idxs > rcap);
|
|
|
|
|
|
|
|
return num;
|
|
|
|
}
|
|
|
|
|
2007-07-18 09:19:09 -04:00
|
|
|
static int
|
|
|
|
ext4_ext_max_entries(struct inode *inode, int depth)
|
|
|
|
{
|
|
|
|
int max;
|
|
|
|
|
|
|
|
if (depth == ext_depth(inode)) {
|
|
|
|
if (depth == 0)
|
|
|
|
max = ext4_ext_space_root(inode);
|
|
|
|
else
|
|
|
|
max = ext4_ext_space_root_idx(inode);
|
|
|
|
} else {
|
|
|
|
if (depth == 0)
|
|
|
|
max = ext4_ext_space_block(inode);
|
|
|
|
else
|
|
|
|
max = ext4_ext_space_block_idx(inode);
|
|
|
|
}
|
|
|
|
|
|
|
|
return max;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int __ext4_ext_check_header(const char *function, struct inode *inode,
|
|
|
|
struct ext4_extent_header *eh,
|
|
|
|
int depth)
|
|
|
|
{
|
|
|
|
const char *error_msg;
|
|
|
|
int max = 0;
|
|
|
|
|
|
|
|
if (unlikely(eh->eh_magic != EXT4_EXT_MAGIC)) {
|
|
|
|
error_msg = "invalid magic";
|
|
|
|
goto corrupted;
|
|
|
|
}
|
|
|
|
if (unlikely(le16_to_cpu(eh->eh_depth) != depth)) {
|
|
|
|
error_msg = "unexpected eh_depth";
|
|
|
|
goto corrupted;
|
|
|
|
}
|
|
|
|
if (unlikely(eh->eh_max == 0)) {
|
|
|
|
error_msg = "invalid eh_max";
|
|
|
|
goto corrupted;
|
|
|
|
}
|
|
|
|
max = ext4_ext_max_entries(inode, depth);
|
|
|
|
if (unlikely(le16_to_cpu(eh->eh_max) > max)) {
|
|
|
|
error_msg = "too large eh_max";
|
|
|
|
goto corrupted;
|
|
|
|
}
|
|
|
|
if (unlikely(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max))) {
|
|
|
|
error_msg = "invalid eh_entries";
|
|
|
|
goto corrupted;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
corrupted:
|
|
|
|
ext4_error(inode->i_sb, function,
|
|
|
|
"bad header in inode #%lu: %s - magic %x, "
|
|
|
|
"entries %u, max %u(%u), depth %u(%u)",
|
|
|
|
inode->i_ino, error_msg, le16_to_cpu(eh->eh_magic),
|
|
|
|
le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max),
|
|
|
|
max, le16_to_cpu(eh->eh_depth), depth);
|
|
|
|
|
|
|
|
return -EIO;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define ext4_ext_check_header(inode, eh, depth) \
|
2008-04-17 10:38:59 -04:00
|
|
|
__ext4_ext_check_header(__func__, inode, eh, depth)
|
2007-07-18 09:19:09 -04:00
|
|
|
|
2006-10-11 04:21:03 -04:00
|
|
|
#ifdef EXT_DEBUG
|
|
|
|
static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path)
|
|
|
|
{
|
|
|
|
int k, l = path->p_depth;
|
|
|
|
|
|
|
|
ext_debug("path:");
|
|
|
|
for (k = 0; k <= l; k++, path++) {
|
|
|
|
if (path->p_idx) {
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug(" %d->%llu", le32_to_cpu(path->p_idx->ei_block),
|
2006-10-11 04:21:05 -04:00
|
|
|
idx_pblock(path->p_idx));
|
2006-10-11 04:21:03 -04:00
|
|
|
} else if (path->p_ext) {
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug(" %d:%d:%llu ",
|
2006-10-11 04:21:03 -04:00
|
|
|
le32_to_cpu(path->p_ext->ee_block),
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_get_actual_len(path->p_ext),
|
2006-10-11 04:21:05 -04:00
|
|
|
ext_pblock(path->p_ext));
|
2006-10-11 04:21:03 -04:00
|
|
|
} else
|
|
|
|
ext_debug(" []");
|
|
|
|
}
|
|
|
|
ext_debug("\n");
|
|
|
|
}
|
|
|
|
|
|
|
|
static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path)
|
|
|
|
{
|
|
|
|
int depth = ext_depth(inode);
|
|
|
|
struct ext4_extent_header *eh;
|
|
|
|
struct ext4_extent *ex;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
if (!path)
|
|
|
|
return;
|
|
|
|
|
|
|
|
eh = path[depth].p_hdr;
|
|
|
|
ex = EXT_FIRST_EXTENT(eh);
|
|
|
|
|
|
|
|
for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ex++) {
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("%d:%d:%llu ", le32_to_cpu(ex->ee_block),
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_get_actual_len(ex), ext_pblock(ex));
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
ext_debug("\n");
|
|
|
|
}
|
|
|
|
#else
|
|
|
|
#define ext4_ext_show_path(inode,path)
|
|
|
|
#define ext4_ext_show_leaf(inode,path)
|
|
|
|
#endif
|
|
|
|
|
2008-02-25 16:54:37 -05:00
|
|
|
void ext4_ext_drop_refs(struct ext4_ext_path *path)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
int depth = path->p_depth;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i <= depth; i++, path++)
|
|
|
|
if (path->p_bh) {
|
|
|
|
brelse(path->p_bh);
|
|
|
|
path->p_bh = NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_binsearch_idx:
|
|
|
|
* binary search for the closest index of the given block
|
2007-07-18 09:19:09 -04:00
|
|
|
* the header must be checked before calling this
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
static void
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_ext_binsearch_idx(struct inode *inode,
|
|
|
|
struct ext4_ext_path *path, ext4_lblk_t block)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
struct ext4_extent_header *eh = path->p_hdr;
|
|
|
|
struct ext4_extent_idx *r, *l, *m;
|
|
|
|
|
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("binsearch for %u(idx): ", block);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
l = EXT_FIRST_INDEX(eh) + 1;
|
2007-07-18 09:09:15 -04:00
|
|
|
r = EXT_LAST_INDEX(eh);
|
2006-10-11 04:21:03 -04:00
|
|
|
while (l <= r) {
|
|
|
|
m = l + (r - l) / 2;
|
|
|
|
if (block < le32_to_cpu(m->ei_block))
|
|
|
|
r = m - 1;
|
|
|
|
else
|
|
|
|
l = m + 1;
|
2007-07-18 08:33:37 -04:00
|
|
|
ext_debug("%p(%u):%p(%u):%p(%u) ", l, le32_to_cpu(l->ei_block),
|
|
|
|
m, le32_to_cpu(m->ei_block),
|
|
|
|
r, le32_to_cpu(r->ei_block));
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
path->p_idx = l - 1;
|
2006-10-11 04:21:05 -04:00
|
|
|
ext_debug(" -> %d->%lld ", le32_to_cpu(path->p_idx->ei_block),
|
2007-07-18 08:33:37 -04:00
|
|
|
idx_pblock(path->p_idx));
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
#ifdef CHECK_BINSEARCH
|
|
|
|
{
|
|
|
|
struct ext4_extent_idx *chix, *ix;
|
|
|
|
int k;
|
|
|
|
|
|
|
|
chix = ix = EXT_FIRST_INDEX(eh);
|
|
|
|
for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ix++) {
|
|
|
|
if (k != 0 &&
|
|
|
|
le32_to_cpu(ix->ei_block) <= le32_to_cpu(ix[-1].ei_block)) {
|
|
|
|
printk("k=%d, ix=0x%p, first=0x%p\n", k,
|
|
|
|
ix, EXT_FIRST_INDEX(eh));
|
|
|
|
printk("%u <= %u\n",
|
|
|
|
le32_to_cpu(ix->ei_block),
|
|
|
|
le32_to_cpu(ix[-1].ei_block));
|
|
|
|
}
|
|
|
|
BUG_ON(k && le32_to_cpu(ix->ei_block)
|
2007-05-24 13:04:54 -04:00
|
|
|
<= le32_to_cpu(ix[-1].ei_block));
|
2006-10-11 04:21:03 -04:00
|
|
|
if (block < le32_to_cpu(ix->ei_block))
|
|
|
|
break;
|
|
|
|
chix = ix;
|
|
|
|
}
|
|
|
|
BUG_ON(chix != path->p_idx);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_binsearch:
|
|
|
|
* binary search for closest extent of the given block
|
2007-07-18 09:19:09 -04:00
|
|
|
* the header must be checked before calling this
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
static void
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_ext_binsearch(struct inode *inode,
|
|
|
|
struct ext4_ext_path *path, ext4_lblk_t block)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
struct ext4_extent_header *eh = path->p_hdr;
|
|
|
|
struct ext4_extent *r, *l, *m;
|
|
|
|
|
|
|
|
if (eh->eh_entries == 0) {
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* this leaf is empty:
|
|
|
|
* we get such a leaf in split/add case
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("binsearch for %u: ", block);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
l = EXT_FIRST_EXTENT(eh) + 1;
|
2007-07-18 09:09:15 -04:00
|
|
|
r = EXT_LAST_EXTENT(eh);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
while (l <= r) {
|
|
|
|
m = l + (r - l) / 2;
|
|
|
|
if (block < le32_to_cpu(m->ee_block))
|
|
|
|
r = m - 1;
|
|
|
|
else
|
|
|
|
l = m + 1;
|
2007-07-18 08:33:37 -04:00
|
|
|
ext_debug("%p(%u):%p(%u):%p(%u) ", l, le32_to_cpu(l->ee_block),
|
|
|
|
m, le32_to_cpu(m->ee_block),
|
|
|
|
r, le32_to_cpu(r->ee_block));
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
path->p_ext = l - 1;
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug(" -> %d:%llu:%d ",
|
2007-05-24 13:04:54 -04:00
|
|
|
le32_to_cpu(path->p_ext->ee_block),
|
|
|
|
ext_pblock(path->p_ext),
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_get_actual_len(path->p_ext));
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
#ifdef CHECK_BINSEARCH
|
|
|
|
{
|
|
|
|
struct ext4_extent *chex, *ex;
|
|
|
|
int k;
|
|
|
|
|
|
|
|
chex = ex = EXT_FIRST_EXTENT(eh);
|
|
|
|
for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ex++) {
|
|
|
|
BUG_ON(k && le32_to_cpu(ex->ee_block)
|
2007-05-24 13:04:54 -04:00
|
|
|
<= le32_to_cpu(ex[-1].ee_block));
|
2006-10-11 04:21:03 -04:00
|
|
|
if (block < le32_to_cpu(ex->ee_block))
|
|
|
|
break;
|
|
|
|
chex = ex;
|
|
|
|
}
|
|
|
|
BUG_ON(chex != path->p_ext);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
int ext4_ext_tree_init(handle_t *handle, struct inode *inode)
|
|
|
|
{
|
|
|
|
struct ext4_extent_header *eh;
|
|
|
|
|
|
|
|
eh = ext_inode_hdr(inode);
|
|
|
|
eh->eh_depth = 0;
|
|
|
|
eh->eh_entries = 0;
|
|
|
|
eh->eh_magic = EXT4_EXT_MAGIC;
|
|
|
|
eh->eh_max = cpu_to_le16(ext4_ext_space_root(inode));
|
|
|
|
ext4_mark_inode_dirty(handle, inode);
|
|
|
|
ext4_ext_invalidate_cache(inode);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
struct ext4_ext_path *
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block,
|
|
|
|
struct ext4_ext_path *path)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
struct ext4_extent_header *eh;
|
|
|
|
struct buffer_head *bh;
|
|
|
|
short int depth, i, ppos = 0, alloc = 0;
|
|
|
|
|
|
|
|
eh = ext_inode_hdr(inode);
|
2007-07-18 09:19:09 -04:00
|
|
|
depth = ext_depth(inode);
|
|
|
|
if (ext4_ext_check_header(inode, eh, depth))
|
2006-10-11 04:21:03 -04:00
|
|
|
return ERR_PTR(-EIO);
|
|
|
|
|
|
|
|
|
|
|
|
/* account possible depth increase */
|
|
|
|
if (!path) {
|
2006-12-06 23:41:35 -05:00
|
|
|
path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 2),
|
2006-10-11 04:21:03 -04:00
|
|
|
GFP_NOFS);
|
|
|
|
if (!path)
|
|
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
alloc = 1;
|
|
|
|
}
|
|
|
|
path[0].p_hdr = eh;
|
2008-07-11 19:27:31 -04:00
|
|
|
path[0].p_bh = NULL;
|
2006-10-11 04:21:03 -04:00
|
|
|
|
2007-07-18 09:19:09 -04:00
|
|
|
i = depth;
|
2006-10-11 04:21:03 -04:00
|
|
|
/* walk through the tree */
|
|
|
|
while (i) {
|
|
|
|
ext_debug("depth %d: num %d, max %d\n",
|
|
|
|
ppos, le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max));
|
2007-07-18 09:19:09 -04:00
|
|
|
|
2006-10-11 04:21:03 -04:00
|
|
|
ext4_ext_binsearch_idx(inode, path + ppos, block);
|
2006-10-11 04:21:05 -04:00
|
|
|
path[ppos].p_block = idx_pblock(path[ppos].p_idx);
|
2006-10-11 04:21:03 -04:00
|
|
|
path[ppos].p_depth = i;
|
|
|
|
path[ppos].p_ext = NULL;
|
|
|
|
|
|
|
|
bh = sb_bread(inode->i_sb, path[ppos].p_block);
|
|
|
|
if (!bh)
|
|
|
|
goto err;
|
|
|
|
|
|
|
|
eh = ext_block_hdr(bh);
|
|
|
|
ppos++;
|
|
|
|
BUG_ON(ppos > depth);
|
|
|
|
path[ppos].p_bh = bh;
|
|
|
|
path[ppos].p_hdr = eh;
|
|
|
|
i--;
|
|
|
|
|
2007-07-18 09:19:09 -04:00
|
|
|
if (ext4_ext_check_header(inode, eh, i))
|
2006-10-11 04:21:03 -04:00
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
path[ppos].p_depth = i;
|
|
|
|
path[ppos].p_ext = NULL;
|
|
|
|
path[ppos].p_idx = NULL;
|
|
|
|
|
|
|
|
/* find extent */
|
|
|
|
ext4_ext_binsearch(inode, path + ppos, block);
|
2008-07-11 19:27:31 -04:00
|
|
|
/* if not an empty leaf */
|
|
|
|
if (path[ppos].p_ext)
|
|
|
|
path[ppos].p_block = ext_pblock(path[ppos].p_ext);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
ext4_ext_show_path(inode, path);
|
|
|
|
|
|
|
|
return path;
|
|
|
|
|
|
|
|
err:
|
|
|
|
ext4_ext_drop_refs(path);
|
|
|
|
if (alloc)
|
|
|
|
kfree(path);
|
|
|
|
return ERR_PTR(-EIO);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_insert_index:
|
|
|
|
* insert new index [@logical;@ptr] into the block at @curp;
|
|
|
|
* check where to insert: before @curp or after @curp
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
static int ext4_ext_insert_index(handle_t *handle, struct inode *inode,
|
|
|
|
struct ext4_ext_path *curp,
|
2006-10-11 04:21:05 -04:00
|
|
|
int logical, ext4_fsblk_t ptr)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
struct ext4_extent_idx *ix;
|
|
|
|
int len, err;
|
|
|
|
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_get_access(handle, inode, curp);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
return err;
|
|
|
|
|
|
|
|
BUG_ON(logical == le32_to_cpu(curp->p_idx->ei_block));
|
|
|
|
len = EXT_MAX_INDEX(curp->p_hdr) - curp->p_idx;
|
|
|
|
if (logical > le32_to_cpu(curp->p_idx->ei_block)) {
|
|
|
|
/* insert after */
|
|
|
|
if (curp->p_idx != EXT_LAST_INDEX(curp->p_hdr)) {
|
|
|
|
len = (len - 1) * sizeof(struct ext4_extent_idx);
|
|
|
|
len = len < 0 ? 0 : len;
|
2007-07-18 08:33:37 -04:00
|
|
|
ext_debug("insert new index %d after: %llu. "
|
2006-10-11 04:21:03 -04:00
|
|
|
"move %d from 0x%p to 0x%p\n",
|
|
|
|
logical, ptr, len,
|
|
|
|
(curp->p_idx + 1), (curp->p_idx + 2));
|
|
|
|
memmove(curp->p_idx + 2, curp->p_idx + 1, len);
|
|
|
|
}
|
|
|
|
ix = curp->p_idx + 1;
|
|
|
|
} else {
|
|
|
|
/* insert before */
|
|
|
|
len = len * sizeof(struct ext4_extent_idx);
|
|
|
|
len = len < 0 ? 0 : len;
|
2007-07-18 08:33:37 -04:00
|
|
|
ext_debug("insert new index %d before: %llu. "
|
2006-10-11 04:21:03 -04:00
|
|
|
"move %d from 0x%p to 0x%p\n",
|
|
|
|
logical, ptr, len,
|
|
|
|
curp->p_idx, (curp->p_idx + 1));
|
|
|
|
memmove(curp->p_idx + 1, curp->p_idx, len);
|
|
|
|
ix = curp->p_idx;
|
|
|
|
}
|
|
|
|
|
|
|
|
ix->ei_block = cpu_to_le32(logical);
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_idx_store_pblock(ix, ptr);
|
2008-04-17 10:38:59 -04:00
|
|
|
le16_add_cpu(&curp->p_hdr->eh_entries, 1);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
BUG_ON(le16_to_cpu(curp->p_hdr->eh_entries)
|
2007-05-24 13:04:54 -04:00
|
|
|
> le16_to_cpu(curp->p_hdr->eh_max));
|
2006-10-11 04:21:03 -04:00
|
|
|
BUG_ON(ix > EXT_LAST_INDEX(curp->p_hdr));
|
|
|
|
|
|
|
|
err = ext4_ext_dirty(handle, inode, curp);
|
|
|
|
ext4_std_error(inode->i_sb, err);
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_split:
|
|
|
|
* inserts new subtree into the path, using free index entry
|
|
|
|
* at depth @at:
|
|
|
|
* - allocates all needed blocks (new leaf and all intermediate index blocks)
|
|
|
|
* - makes decision where to split
|
|
|
|
* - moves remaining extents and index entries (right to the split point)
|
|
|
|
* into the newly allocated blocks
|
|
|
|
* - initializes subtree
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
static int ext4_ext_split(handle_t *handle, struct inode *inode,
|
|
|
|
struct ext4_ext_path *path,
|
|
|
|
struct ext4_extent *newext, int at)
|
|
|
|
{
|
|
|
|
struct buffer_head *bh = NULL;
|
|
|
|
int depth = ext_depth(inode);
|
|
|
|
struct ext4_extent_header *neh;
|
|
|
|
struct ext4_extent_idx *fidx;
|
|
|
|
struct ext4_extent *ex;
|
|
|
|
int i = at, k, m, a;
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_fsblk_t newblock, oldblock;
|
2006-10-11 04:21:03 -04:00
|
|
|
__le32 border;
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_fsblk_t *ablocks = NULL; /* array of allocated blocks */
|
2006-10-11 04:21:03 -04:00
|
|
|
int err = 0;
|
|
|
|
|
|
|
|
/* make decision: where to split? */
|
2006-10-11 04:21:07 -04:00
|
|
|
/* FIXME: now decision is simplest: at current extent */
|
2006-10-11 04:21:03 -04:00
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/* if current leaf will be split, then we should use
|
2006-10-11 04:21:03 -04:00
|
|
|
* border from split point */
|
|
|
|
BUG_ON(path[depth].p_ext > EXT_MAX_EXTENT(path[depth].p_hdr));
|
|
|
|
if (path[depth].p_ext != EXT_MAX_EXTENT(path[depth].p_hdr)) {
|
|
|
|
border = path[depth].p_ext[1].ee_block;
|
2006-10-11 04:21:07 -04:00
|
|
|
ext_debug("leaf will be split."
|
2006-10-11 04:21:03 -04:00
|
|
|
" next leaf starts at %d\n",
|
2007-05-24 13:04:54 -04:00
|
|
|
le32_to_cpu(border));
|
2006-10-11 04:21:03 -04:00
|
|
|
} else {
|
|
|
|
border = newext->ee_block;
|
|
|
|
ext_debug("leaf will be added."
|
|
|
|
" next leaf starts at %d\n",
|
2007-05-24 13:04:54 -04:00
|
|
|
le32_to_cpu(border));
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* If error occurs, then we break processing
|
|
|
|
* and mark filesystem read-only. index won't
|
2006-10-11 04:21:03 -04:00
|
|
|
* be inserted and tree will be in consistent
|
2006-10-11 04:21:07 -04:00
|
|
|
* state. Next mount will repair buffers too.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* Get array to track all allocated blocks.
|
|
|
|
* We need this to handle errors and free blocks
|
|
|
|
* upon them.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
2006-12-06 23:41:35 -05:00
|
|
|
ablocks = kzalloc(sizeof(ext4_fsblk_t) * depth, GFP_NOFS);
|
2006-10-11 04:21:03 -04:00
|
|
|
if (!ablocks)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
/* allocate all needed blocks */
|
|
|
|
ext_debug("allocate %d blocks for indexes/leaf\n", depth - at);
|
|
|
|
for (a = 0; a < depth - at; a++) {
|
2008-07-11 19:27:31 -04:00
|
|
|
newblock = ext4_ext_new_meta_block(handle, inode, path,
|
|
|
|
newext, &err);
|
2006-10-11 04:21:03 -04:00
|
|
|
if (newblock == 0)
|
|
|
|
goto cleanup;
|
|
|
|
ablocks[a] = newblock;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* initialize new leaf */
|
|
|
|
newblock = ablocks[--a];
|
|
|
|
BUG_ON(newblock == 0);
|
|
|
|
bh = sb_getblk(inode->i_sb, newblock);
|
|
|
|
if (!bh) {
|
|
|
|
err = -EIO;
|
|
|
|
goto cleanup;
|
|
|
|
}
|
|
|
|
lock_buffer(bh);
|
|
|
|
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_journal_get_create_access(handle, bh);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
goto cleanup;
|
|
|
|
|
|
|
|
neh = ext_block_hdr(bh);
|
|
|
|
neh->eh_entries = 0;
|
|
|
|
neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode));
|
|
|
|
neh->eh_magic = EXT4_EXT_MAGIC;
|
|
|
|
neh->eh_depth = 0;
|
|
|
|
ex = EXT_FIRST_EXTENT(neh);
|
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/* move remainder of path[depth] to the new leaf */
|
2006-10-11 04:21:03 -04:00
|
|
|
BUG_ON(path[depth].p_hdr->eh_entries != path[depth].p_hdr->eh_max);
|
|
|
|
/* start copy from next extent */
|
|
|
|
/* TODO: we could do it by single memmove */
|
|
|
|
m = 0;
|
|
|
|
path[depth].p_ext++;
|
|
|
|
while (path[depth].p_ext <=
|
|
|
|
EXT_MAX_EXTENT(path[depth].p_hdr)) {
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("move %d:%llu:%d in new leaf %llu\n",
|
2007-05-24 13:04:54 -04:00
|
|
|
le32_to_cpu(path[depth].p_ext->ee_block),
|
|
|
|
ext_pblock(path[depth].p_ext),
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_get_actual_len(path[depth].p_ext),
|
2006-10-11 04:21:03 -04:00
|
|
|
newblock);
|
|
|
|
/*memmove(ex++, path[depth].p_ext++,
|
|
|
|
sizeof(struct ext4_extent));
|
|
|
|
neh->eh_entries++;*/
|
|
|
|
path[depth].p_ext++;
|
|
|
|
m++;
|
|
|
|
}
|
|
|
|
if (m) {
|
|
|
|
memmove(ex, path[depth].p_ext-m, sizeof(struct ext4_extent)*m);
|
2008-04-17 10:38:59 -04:00
|
|
|
le16_add_cpu(&neh->eh_entries, m);
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
set_buffer_uptodate(bh);
|
|
|
|
unlock_buffer(bh);
|
|
|
|
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_journal_dirty_metadata(handle, bh);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
goto cleanup;
|
|
|
|
brelse(bh);
|
|
|
|
bh = NULL;
|
|
|
|
|
|
|
|
/* correct old leaf */
|
|
|
|
if (m) {
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_get_access(handle, inode, path + depth);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
goto cleanup;
|
2008-04-17 10:38:59 -04:00
|
|
|
le16_add_cpu(&path[depth].p_hdr->eh_entries, -m);
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_dirty(handle, inode, path + depth);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
goto cleanup;
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
/* create intermediate indexes */
|
|
|
|
k = depth - at - 1;
|
|
|
|
BUG_ON(k < 0);
|
|
|
|
if (k)
|
|
|
|
ext_debug("create %d intermediate indices\n", k);
|
|
|
|
/* insert new index into current index block */
|
|
|
|
/* current depth stored in i var */
|
|
|
|
i = depth - 1;
|
|
|
|
while (k--) {
|
|
|
|
oldblock = newblock;
|
|
|
|
newblock = ablocks[--a];
|
2008-01-28 23:58:27 -05:00
|
|
|
bh = sb_getblk(inode->i_sb, newblock);
|
2006-10-11 04:21:03 -04:00
|
|
|
if (!bh) {
|
|
|
|
err = -EIO;
|
|
|
|
goto cleanup;
|
|
|
|
}
|
|
|
|
lock_buffer(bh);
|
|
|
|
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_journal_get_create_access(handle, bh);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
goto cleanup;
|
|
|
|
|
|
|
|
neh = ext_block_hdr(bh);
|
|
|
|
neh->eh_entries = cpu_to_le16(1);
|
|
|
|
neh->eh_magic = EXT4_EXT_MAGIC;
|
|
|
|
neh->eh_max = cpu_to_le16(ext4_ext_space_block_idx(inode));
|
|
|
|
neh->eh_depth = cpu_to_le16(depth - i);
|
|
|
|
fidx = EXT_FIRST_INDEX(neh);
|
|
|
|
fidx->ei_block = border;
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_idx_store_pblock(fidx, oldblock);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("int.index at %d (block %llu): %u -> %llu\n",
|
|
|
|
i, newblock, le32_to_cpu(border), oldblock);
|
2006-10-11 04:21:03 -04:00
|
|
|
/* copy indexes */
|
|
|
|
m = 0;
|
|
|
|
path[i].p_idx++;
|
|
|
|
|
|
|
|
ext_debug("cur 0x%p, last 0x%p\n", path[i].p_idx,
|
|
|
|
EXT_MAX_INDEX(path[i].p_hdr));
|
|
|
|
BUG_ON(EXT_MAX_INDEX(path[i].p_hdr) !=
|
|
|
|
EXT_LAST_INDEX(path[i].p_hdr));
|
|
|
|
while (path[i].p_idx <= EXT_MAX_INDEX(path[i].p_hdr)) {
|
2007-07-18 08:33:37 -04:00
|
|
|
ext_debug("%d: move %d:%llu in new index %llu\n", i,
|
2007-05-24 13:04:54 -04:00
|
|
|
le32_to_cpu(path[i].p_idx->ei_block),
|
|
|
|
idx_pblock(path[i].p_idx),
|
|
|
|
newblock);
|
2006-10-11 04:21:03 -04:00
|
|
|
/*memmove(++fidx, path[i].p_idx++,
|
|
|
|
sizeof(struct ext4_extent_idx));
|
|
|
|
neh->eh_entries++;
|
|
|
|
BUG_ON(neh->eh_entries > neh->eh_max);*/
|
|
|
|
path[i].p_idx++;
|
|
|
|
m++;
|
|
|
|
}
|
|
|
|
if (m) {
|
|
|
|
memmove(++fidx, path[i].p_idx - m,
|
|
|
|
sizeof(struct ext4_extent_idx) * m);
|
2008-04-17 10:38:59 -04:00
|
|
|
le16_add_cpu(&neh->eh_entries, m);
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
set_buffer_uptodate(bh);
|
|
|
|
unlock_buffer(bh);
|
|
|
|
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_journal_dirty_metadata(handle, bh);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
goto cleanup;
|
|
|
|
brelse(bh);
|
|
|
|
bh = NULL;
|
|
|
|
|
|
|
|
/* correct old index */
|
|
|
|
if (m) {
|
|
|
|
err = ext4_ext_get_access(handle, inode, path + i);
|
|
|
|
if (err)
|
|
|
|
goto cleanup;
|
2008-04-17 10:38:59 -04:00
|
|
|
le16_add_cpu(&path[i].p_hdr->eh_entries, -m);
|
2006-10-11 04:21:03 -04:00
|
|
|
err = ext4_ext_dirty(handle, inode, path + i);
|
|
|
|
if (err)
|
|
|
|
goto cleanup;
|
|
|
|
}
|
|
|
|
|
|
|
|
i--;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* insert new index */
|
|
|
|
err = ext4_ext_insert_index(handle, inode, path + at,
|
|
|
|
le32_to_cpu(border), newblock);
|
|
|
|
|
|
|
|
cleanup:
|
|
|
|
if (bh) {
|
|
|
|
if (buffer_locked(bh))
|
|
|
|
unlock_buffer(bh);
|
|
|
|
brelse(bh);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (err) {
|
|
|
|
/* free all allocated blocks in error case */
|
|
|
|
for (i = 0; i < depth; i++) {
|
|
|
|
if (!ablocks[i])
|
|
|
|
continue;
|
2008-01-29 00:19:52 -05:00
|
|
|
ext4_free_blocks(handle, inode, ablocks[i], 1, 1);
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
kfree(ablocks);
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_grow_indepth:
|
|
|
|
* implements tree growing procedure:
|
|
|
|
* - allocates new block
|
|
|
|
* - moves top-level data (index block or leaf) into the new block
|
|
|
|
* - initializes new top-level, creating index that points to the
|
|
|
|
* just created block
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode,
|
|
|
|
struct ext4_ext_path *path,
|
|
|
|
struct ext4_extent *newext)
|
|
|
|
{
|
|
|
|
struct ext4_ext_path *curp = path;
|
|
|
|
struct ext4_extent_header *neh;
|
|
|
|
struct ext4_extent_idx *fidx;
|
|
|
|
struct buffer_head *bh;
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_fsblk_t newblock;
|
2006-10-11 04:21:03 -04:00
|
|
|
int err = 0;
|
|
|
|
|
2008-07-11 19:27:31 -04:00
|
|
|
newblock = ext4_ext_new_meta_block(handle, inode, path, newext, &err);
|
2006-10-11 04:21:03 -04:00
|
|
|
if (newblock == 0)
|
|
|
|
return err;
|
|
|
|
|
|
|
|
bh = sb_getblk(inode->i_sb, newblock);
|
|
|
|
if (!bh) {
|
|
|
|
err = -EIO;
|
|
|
|
ext4_std_error(inode->i_sb, err);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
lock_buffer(bh);
|
|
|
|
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_journal_get_create_access(handle, bh);
|
|
|
|
if (err) {
|
2006-10-11 04:21:03 -04:00
|
|
|
unlock_buffer(bh);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* move top-level index/leaf into new block */
|
|
|
|
memmove(bh->b_data, curp->p_hdr, sizeof(EXT4_I(inode)->i_data));
|
|
|
|
|
|
|
|
/* set size of new block */
|
|
|
|
neh = ext_block_hdr(bh);
|
|
|
|
/* old root could have indexes or leaves
|
|
|
|
* so calculate e_max right way */
|
|
|
|
if (ext_depth(inode))
|
|
|
|
neh->eh_max = cpu_to_le16(ext4_ext_space_block_idx(inode));
|
|
|
|
else
|
|
|
|
neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode));
|
|
|
|
neh->eh_magic = EXT4_EXT_MAGIC;
|
|
|
|
set_buffer_uptodate(bh);
|
|
|
|
unlock_buffer(bh);
|
|
|
|
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_journal_dirty_metadata(handle, bh);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
goto out;
|
|
|
|
|
|
|
|
/* create index in new top-level index: num,max,pointer */
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_get_access(handle, inode, curp);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
goto out;
|
|
|
|
|
|
|
|
curp->p_hdr->eh_magic = EXT4_EXT_MAGIC;
|
|
|
|
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);
|
2007-07-18 09:09:15 -04:00
|
|
|
|
|
|
|
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;
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_idx_store_pblock(curp->p_idx, newblock);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
neh = ext_inode_hdr(inode);
|
|
|
|
fidx = EXT_FIRST_INDEX(neh);
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("new root: num %d(%d), lblock %d, ptr %llu\n",
|
2006-10-11 04:21:03 -04:00
|
|
|
le16_to_cpu(neh->eh_entries), le16_to_cpu(neh->eh_max),
|
2006-10-11 04:21:05 -04:00
|
|
|
le32_to_cpu(fidx->ei_block), idx_pblock(fidx));
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
neh->eh_depth = cpu_to_le16(path->p_depth + 1);
|
|
|
|
err = ext4_ext_dirty(handle, inode, curp);
|
|
|
|
out:
|
|
|
|
brelse(bh);
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_create_new_leaf:
|
|
|
|
* finds empty index and adds new leaf.
|
|
|
|
* if no free index is found, then it requests in-depth growing.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
static int ext4_ext_create_new_leaf(handle_t *handle, struct inode *inode,
|
|
|
|
struct ext4_ext_path *path,
|
|
|
|
struct ext4_extent *newext)
|
|
|
|
{
|
|
|
|
struct ext4_ext_path *curp;
|
|
|
|
int depth, i, err = 0;
|
|
|
|
|
|
|
|
repeat:
|
|
|
|
i = depth = ext_depth(inode);
|
|
|
|
|
|
|
|
/* walk up to the tree and look for free index entry */
|
|
|
|
curp = path + depth;
|
|
|
|
while (i > 0 && !EXT_HAS_FREE_INDEX(curp)) {
|
|
|
|
i--;
|
|
|
|
curp--;
|
|
|
|
}
|
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/* we use already allocated block for index block,
|
|
|
|
* so subsequent data blocks should be contiguous */
|
2006-10-11 04:21:03 -04:00
|
|
|
if (EXT_HAS_FREE_INDEX(curp)) {
|
|
|
|
/* if we found index with free entry, then use that
|
|
|
|
* entry: create all needed subtree and add new leaf */
|
|
|
|
err = ext4_ext_split(handle, inode, path, newext, i);
|
2008-07-11 19:27:31 -04:00
|
|
|
if (err)
|
|
|
|
goto out;
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
/* refill path */
|
|
|
|
ext4_ext_drop_refs(path);
|
|
|
|
path = ext4_ext_find_extent(inode,
|
2008-01-28 23:58:27 -05:00
|
|
|
(ext4_lblk_t)le32_to_cpu(newext->ee_block),
|
|
|
|
path);
|
2006-10-11 04:21:03 -04:00
|
|
|
if (IS_ERR(path))
|
|
|
|
err = PTR_ERR(path);
|
|
|
|
} else {
|
|
|
|
/* tree is full, time to grow in depth */
|
|
|
|
err = ext4_ext_grow_indepth(handle, inode, path, newext);
|
|
|
|
if (err)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
/* refill path */
|
|
|
|
ext4_ext_drop_refs(path);
|
|
|
|
path = ext4_ext_find_extent(inode,
|
2008-01-28 23:58:27 -05:00
|
|
|
(ext4_lblk_t)le32_to_cpu(newext->ee_block),
|
|
|
|
path);
|
2006-10-11 04:21:03 -04:00
|
|
|
if (IS_ERR(path)) {
|
|
|
|
err = PTR_ERR(path);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* only first (depth 0 -> 1) produces free space;
|
|
|
|
* in all other cases we have to split the grown tree
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
depth = ext_depth(inode);
|
|
|
|
if (path[depth].p_hdr->eh_entries == path[depth].p_hdr->eh_max) {
|
2006-10-11 04:21:07 -04:00
|
|
|
/* now we need to split */
|
2006-10-11 04:21:03 -04:00
|
|
|
goto repeat;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
out:
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
/*
|
|
|
|
* search the closest allocated block to the left for *logical
|
|
|
|
* and returns it at @logical + it's physical address at @phys
|
|
|
|
* if *logical is the smallest allocated block, the function
|
|
|
|
* returns 0 at @phys
|
|
|
|
* return value contains 0 (success) or error code
|
|
|
|
*/
|
|
|
|
int
|
|
|
|
ext4_ext_search_left(struct inode *inode, struct ext4_ext_path *path,
|
|
|
|
ext4_lblk_t *logical, ext4_fsblk_t *phys)
|
|
|
|
{
|
|
|
|
struct ext4_extent_idx *ix;
|
|
|
|
struct ext4_extent *ex;
|
2008-01-28 23:58:27 -05:00
|
|
|
int depth, ee_len;
|
2008-01-28 23:58:27 -05:00
|
|
|
|
|
|
|
BUG_ON(path == NULL);
|
|
|
|
depth = path->p_depth;
|
|
|
|
*phys = 0;
|
|
|
|
|
|
|
|
if (depth == 0 && path->p_ext == NULL)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
/* usually extent in the path covers blocks smaller
|
|
|
|
* then *logical, but it can be that extent is the
|
|
|
|
* first one in the file */
|
|
|
|
|
|
|
|
ex = path[depth].p_ext;
|
2008-01-28 23:58:27 -05:00
|
|
|
ee_len = ext4_ext_get_actual_len(ex);
|
2008-01-28 23:58:27 -05:00
|
|
|
if (*logical < le32_to_cpu(ex->ee_block)) {
|
|
|
|
BUG_ON(EXT_FIRST_EXTENT(path[depth].p_hdr) != ex);
|
|
|
|
while (--depth >= 0) {
|
|
|
|
ix = path[depth].p_idx;
|
|
|
|
BUG_ON(ix != EXT_FIRST_INDEX(path[depth].p_hdr));
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
BUG_ON(*logical < (le32_to_cpu(ex->ee_block) + ee_len));
|
2008-01-28 23:58:27 -05:00
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
*logical = le32_to_cpu(ex->ee_block) + ee_len - 1;
|
|
|
|
*phys = ext_pblock(ex) + ee_len - 1;
|
2008-01-28 23:58:27 -05:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* search the closest allocated block to the right for *logical
|
|
|
|
* and returns it at @logical + it's physical address at @phys
|
|
|
|
* if *logical is the smallest allocated block, the function
|
|
|
|
* returns 0 at @phys
|
|
|
|
* return value contains 0 (success) or error code
|
|
|
|
*/
|
|
|
|
int
|
|
|
|
ext4_ext_search_right(struct inode *inode, struct ext4_ext_path *path,
|
|
|
|
ext4_lblk_t *logical, ext4_fsblk_t *phys)
|
|
|
|
{
|
|
|
|
struct buffer_head *bh = NULL;
|
|
|
|
struct ext4_extent_header *eh;
|
|
|
|
struct ext4_extent_idx *ix;
|
|
|
|
struct ext4_extent *ex;
|
|
|
|
ext4_fsblk_t block;
|
2008-01-28 23:58:27 -05:00
|
|
|
int depth, ee_len;
|
2008-01-28 23:58:27 -05:00
|
|
|
|
|
|
|
BUG_ON(path == NULL);
|
|
|
|
depth = path->p_depth;
|
|
|
|
*phys = 0;
|
|
|
|
|
|
|
|
if (depth == 0 && path->p_ext == NULL)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
/* usually extent in the path covers blocks smaller
|
|
|
|
* then *logical, but it can be that extent is the
|
|
|
|
* first one in the file */
|
|
|
|
|
|
|
|
ex = path[depth].p_ext;
|
2008-01-28 23:58:27 -05:00
|
|
|
ee_len = ext4_ext_get_actual_len(ex);
|
2008-01-28 23:58:27 -05:00
|
|
|
if (*logical < le32_to_cpu(ex->ee_block)) {
|
|
|
|
BUG_ON(EXT_FIRST_EXTENT(path[depth].p_hdr) != ex);
|
|
|
|
while (--depth >= 0) {
|
|
|
|
ix = path[depth].p_idx;
|
|
|
|
BUG_ON(ix != EXT_FIRST_INDEX(path[depth].p_hdr));
|
|
|
|
}
|
|
|
|
*logical = le32_to_cpu(ex->ee_block);
|
|
|
|
*phys = ext_pblock(ex);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
BUG_ON(*logical < (le32_to_cpu(ex->ee_block) + ee_len));
|
2008-01-28 23:58:27 -05:00
|
|
|
|
|
|
|
if (ex != EXT_LAST_EXTENT(path[depth].p_hdr)) {
|
|
|
|
/* next allocated block in this leaf */
|
|
|
|
ex++;
|
|
|
|
*logical = le32_to_cpu(ex->ee_block);
|
|
|
|
*phys = ext_pblock(ex);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* go up and search for index to the right */
|
|
|
|
while (--depth >= 0) {
|
|
|
|
ix = path[depth].p_idx;
|
|
|
|
if (ix != EXT_LAST_INDEX(path[depth].p_hdr))
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (depth < 0) {
|
|
|
|
/* we've gone up to the root and
|
|
|
|
* found no index to the right */
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* we've found index to the right, let's
|
|
|
|
* follow it and find the closest allocated
|
|
|
|
* block to the right */
|
|
|
|
ix++;
|
|
|
|
block = idx_pblock(ix);
|
|
|
|
while (++depth < path->p_depth) {
|
|
|
|
bh = sb_bread(inode->i_sb, block);
|
|
|
|
if (bh == NULL)
|
|
|
|
return -EIO;
|
|
|
|
eh = ext_block_hdr(bh);
|
|
|
|
if (ext4_ext_check_header(inode, eh, depth)) {
|
|
|
|
put_bh(bh);
|
|
|
|
return -EIO;
|
|
|
|
}
|
|
|
|
ix = EXT_FIRST_INDEX(eh);
|
|
|
|
block = idx_pblock(ix);
|
|
|
|
put_bh(bh);
|
|
|
|
}
|
|
|
|
|
|
|
|
bh = sb_bread(inode->i_sb, block);
|
|
|
|
if (bh == NULL)
|
|
|
|
return -EIO;
|
|
|
|
eh = ext_block_hdr(bh);
|
|
|
|
if (ext4_ext_check_header(inode, eh, path->p_depth - depth)) {
|
|
|
|
put_bh(bh);
|
|
|
|
return -EIO;
|
|
|
|
}
|
|
|
|
ex = EXT_FIRST_EXTENT(eh);
|
|
|
|
*logical = le32_to_cpu(ex->ee_block);
|
|
|
|
*phys = ext_pblock(ex);
|
|
|
|
put_bh(bh);
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
}
|
|
|
|
|
2006-10-11 04:21:03 -04:00
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_next_allocated_block:
|
|
|
|
* returns allocated block in subsequent extent or EXT_MAX_BLOCK.
|
|
|
|
* NOTE: it considers block number from index entry as
|
|
|
|
* allocated block. Thus, index entries have to be consistent
|
|
|
|
* with leaves.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
2008-01-28 23:58:27 -05:00
|
|
|
static ext4_lblk_t
|
2006-10-11 04:21:03 -04:00
|
|
|
ext4_ext_next_allocated_block(struct ext4_ext_path *path)
|
|
|
|
{
|
|
|
|
int depth;
|
|
|
|
|
|
|
|
BUG_ON(path == NULL);
|
|
|
|
depth = path->p_depth;
|
|
|
|
|
|
|
|
if (depth == 0 && path->p_ext == NULL)
|
|
|
|
return EXT_MAX_BLOCK;
|
|
|
|
|
|
|
|
while (depth >= 0) {
|
|
|
|
if (depth == path->p_depth) {
|
|
|
|
/* leaf */
|
|
|
|
if (path[depth].p_ext !=
|
|
|
|
EXT_LAST_EXTENT(path[depth].p_hdr))
|
|
|
|
return le32_to_cpu(path[depth].p_ext[1].ee_block);
|
|
|
|
} else {
|
|
|
|
/* index */
|
|
|
|
if (path[depth].p_idx !=
|
|
|
|
EXT_LAST_INDEX(path[depth].p_hdr))
|
|
|
|
return le32_to_cpu(path[depth].p_idx[1].ei_block);
|
|
|
|
}
|
|
|
|
depth--;
|
|
|
|
}
|
|
|
|
|
|
|
|
return EXT_MAX_BLOCK;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_next_leaf_block:
|
2006-10-11 04:21:03 -04:00
|
|
|
* returns first allocated block from next leaf or EXT_MAX_BLOCK
|
|
|
|
*/
|
2008-01-28 23:58:27 -05:00
|
|
|
static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode,
|
2006-10-11 04:21:24 -04:00
|
|
|
struct ext4_ext_path *path)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
int depth;
|
|
|
|
|
|
|
|
BUG_ON(path == NULL);
|
|
|
|
depth = path->p_depth;
|
|
|
|
|
|
|
|
/* zero-tree has no leaf blocks at all */
|
|
|
|
if (depth == 0)
|
|
|
|
return EXT_MAX_BLOCK;
|
|
|
|
|
|
|
|
/* go to index block */
|
|
|
|
depth--;
|
|
|
|
|
|
|
|
while (depth >= 0) {
|
|
|
|
if (path[depth].p_idx !=
|
|
|
|
EXT_LAST_INDEX(path[depth].p_hdr))
|
2008-01-28 23:58:27 -05:00
|
|
|
return (ext4_lblk_t)
|
|
|
|
le32_to_cpu(path[depth].p_idx[1].ei_block);
|
2006-10-11 04:21:03 -04:00
|
|
|
depth--;
|
|
|
|
}
|
|
|
|
|
|
|
|
return EXT_MAX_BLOCK;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_correct_indexes:
|
|
|
|
* if leaf gets modified and modified extent is first in the leaf,
|
|
|
|
* then we have to correct all indexes above.
|
2006-10-11 04:21:03 -04:00
|
|
|
* TODO: do we need to correct tree in all cases?
|
|
|
|
*/
|
2008-01-28 23:58:27 -05:00
|
|
|
static int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode,
|
2006-10-11 04:21:03 -04:00
|
|
|
struct ext4_ext_path *path)
|
|
|
|
{
|
|
|
|
struct ext4_extent_header *eh;
|
|
|
|
int depth = ext_depth(inode);
|
|
|
|
struct ext4_extent *ex;
|
|
|
|
__le32 border;
|
|
|
|
int k, err = 0;
|
|
|
|
|
|
|
|
eh = path[depth].p_hdr;
|
|
|
|
ex = path[depth].p_ext;
|
|
|
|
BUG_ON(ex == NULL);
|
|
|
|
BUG_ON(eh == NULL);
|
|
|
|
|
|
|
|
if (depth == 0) {
|
|
|
|
/* there is no tree at all */
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ex != EXT_FIRST_EXTENT(eh)) {
|
|
|
|
/* we correct tree if first leaf got modified only */
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* TODO: we need correction if border is smaller than current one
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
k = depth - 1;
|
|
|
|
border = path[depth].p_ext->ee_block;
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_get_access(handle, inode, path + k);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
return err;
|
|
|
|
path[k].p_idx->ei_block = border;
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_dirty(handle, inode, path + k);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
return err;
|
|
|
|
|
|
|
|
while (k--) {
|
|
|
|
/* change all left-side indexes */
|
|
|
|
if (path[k+1].p_idx != EXT_FIRST_INDEX(path[k+1].p_hdr))
|
|
|
|
break;
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_get_access(handle, inode, path + k);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
break;
|
|
|
|
path[k].p_idx->ei_block = border;
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_dirty(handle, inode, path + k);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2006-12-06 23:41:36 -05:00
|
|
|
static int
|
2006-10-11 04:21:03 -04:00
|
|
|
ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1,
|
|
|
|
struct ext4_extent *ex2)
|
|
|
|
{
|
2007-07-18 09:02:56 -04:00
|
|
|
unsigned short ext1_ee_len, ext2_ee_len, max_len;
|
2007-07-17 21:42:41 -04:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Make sure that either both extents are uninitialized, or
|
|
|
|
* both are _not_.
|
|
|
|
*/
|
|
|
|
if (ext4_ext_is_uninitialized(ex1) ^ ext4_ext_is_uninitialized(ex2))
|
|
|
|
return 0;
|
|
|
|
|
2007-07-18 09:02:56 -04:00
|
|
|
if (ext4_ext_is_uninitialized(ex1))
|
|
|
|
max_len = EXT_UNINIT_MAX_LEN;
|
|
|
|
else
|
|
|
|
max_len = EXT_INIT_MAX_LEN;
|
|
|
|
|
2007-07-17 21:42:41 -04:00
|
|
|
ext1_ee_len = ext4_ext_get_actual_len(ex1);
|
|
|
|
ext2_ee_len = ext4_ext_get_actual_len(ex2);
|
|
|
|
|
|
|
|
if (le32_to_cpu(ex1->ee_block) + ext1_ee_len !=
|
2006-10-11 04:21:24 -04:00
|
|
|
le32_to_cpu(ex2->ee_block))
|
2006-10-11 04:21:03 -04:00
|
|
|
return 0;
|
|
|
|
|
2006-10-11 04:21:06 -04:00
|
|
|
/*
|
|
|
|
* To allow future support for preallocated extents to be added
|
|
|
|
* as an RO_COMPAT feature, refuse to merge to extents if
|
2006-10-11 04:21:07 -04:00
|
|
|
* this can result in the top bit of ee_len being set.
|
2006-10-11 04:21:06 -04:00
|
|
|
*/
|
2007-07-18 09:02:56 -04:00
|
|
|
if (ext1_ee_len + ext2_ee_len > max_len)
|
2006-10-11 04:21:06 -04:00
|
|
|
return 0;
|
2007-02-17 13:20:16 -05:00
|
|
|
#ifdef AGGRESSIVE_TEST
|
2008-01-28 23:58:27 -05:00
|
|
|
if (ext1_ee_len >= 4)
|
2006-10-11 04:21:03 -04:00
|
|
|
return 0;
|
|
|
|
#endif
|
|
|
|
|
2007-07-17 21:42:41 -04:00
|
|
|
if (ext_pblock(ex1) + ext1_ee_len == ext_pblock(ex2))
|
2006-10-11 04:21:03 -04:00
|
|
|
return 1;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-07-17 21:42:38 -04:00
|
|
|
/*
|
|
|
|
* This function tries to merge the "ex" extent to the next extent in the tree.
|
|
|
|
* It always tries to merge towards right. If you want to merge towards
|
|
|
|
* left, pass "ex - 1" as argument instead of "ex".
|
|
|
|
* Returns 0 if the extents (ex and ex+1) were _not_ merged and returns
|
|
|
|
* 1 if they got merged.
|
|
|
|
*/
|
|
|
|
int ext4_ext_try_to_merge(struct inode *inode,
|
|
|
|
struct ext4_ext_path *path,
|
|
|
|
struct ext4_extent *ex)
|
|
|
|
{
|
|
|
|
struct ext4_extent_header *eh;
|
|
|
|
unsigned int depth, len;
|
|
|
|
int merge_done = 0;
|
|
|
|
int uninitialized = 0;
|
|
|
|
|
|
|
|
depth = ext_depth(inode);
|
|
|
|
BUG_ON(path[depth].p_hdr == NULL);
|
|
|
|
eh = path[depth].p_hdr;
|
|
|
|
|
|
|
|
while (ex < EXT_LAST_EXTENT(eh)) {
|
|
|
|
if (!ext4_can_extents_be_merged(inode, ex, ex + 1))
|
|
|
|
break;
|
|
|
|
/* merge with next extent! */
|
|
|
|
if (ext4_ext_is_uninitialized(ex))
|
|
|
|
uninitialized = 1;
|
|
|
|
ex->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex)
|
|
|
|
+ ext4_ext_get_actual_len(ex + 1));
|
|
|
|
if (uninitialized)
|
|
|
|
ext4_ext_mark_uninitialized(ex);
|
|
|
|
|
|
|
|
if (ex + 1 < EXT_LAST_EXTENT(eh)) {
|
|
|
|
len = (EXT_LAST_EXTENT(eh) - ex - 1)
|
|
|
|
* sizeof(struct ext4_extent);
|
|
|
|
memmove(ex + 1, ex + 2, len);
|
|
|
|
}
|
2008-04-17 10:38:59 -04:00
|
|
|
le16_add_cpu(&eh->eh_entries, -1);
|
2007-07-17 21:42:38 -04:00
|
|
|
merge_done = 1;
|
|
|
|
WARN_ON(eh->eh_entries == 0);
|
|
|
|
if (!eh->eh_entries)
|
|
|
|
ext4_error(inode->i_sb, "ext4_ext_try_to_merge",
|
|
|
|
"inode#%lu, eh->eh_entries = 0!", inode->i_ino);
|
|
|
|
}
|
|
|
|
|
|
|
|
return merge_done;
|
|
|
|
}
|
|
|
|
|
2007-05-24 13:04:13 -04:00
|
|
|
/*
|
|
|
|
* check if a portion of the "newext" extent overlaps with an
|
|
|
|
* existing extent.
|
|
|
|
*
|
|
|
|
* If there is an overlap discovered, it updates the length of the newext
|
|
|
|
* such that there will be no overlap, and then returns 1.
|
|
|
|
* If there is no overlap found, it returns 0.
|
|
|
|
*/
|
|
|
|
unsigned int ext4_ext_check_overlap(struct inode *inode,
|
|
|
|
struct ext4_extent *newext,
|
|
|
|
struct ext4_ext_path *path)
|
|
|
|
{
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t b1, b2;
|
2007-05-24 13:04:13 -04:00
|
|
|
unsigned int depth, len1;
|
|
|
|
unsigned int ret = 0;
|
|
|
|
|
|
|
|
b1 = le32_to_cpu(newext->ee_block);
|
2007-07-17 21:42:41 -04:00
|
|
|
len1 = ext4_ext_get_actual_len(newext);
|
2007-05-24 13:04:13 -04:00
|
|
|
depth = ext_depth(inode);
|
|
|
|
if (!path[depth].p_ext)
|
|
|
|
goto out;
|
|
|
|
b2 = le32_to_cpu(path[depth].p_ext->ee_block);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* get the next allocated block if the extent in the path
|
|
|
|
* is before the requested block(s)
|
|
|
|
*/
|
|
|
|
if (b2 < b1) {
|
|
|
|
b2 = ext4_ext_next_allocated_block(path);
|
|
|
|
if (b2 == EXT_MAX_BLOCK)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
/* check for wrap through zero on extent logical start block*/
|
2007-05-24 13:04:13 -04:00
|
|
|
if (b1 + len1 < b1) {
|
|
|
|
len1 = EXT_MAX_BLOCK - b1;
|
|
|
|
newext->ee_len = cpu_to_le16(len1);
|
|
|
|
ret = 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* check for overlap */
|
|
|
|
if (b1 + len1 > b2) {
|
|
|
|
newext->ee_len = cpu_to_le16(b2 - b1);
|
|
|
|
ret = 1;
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2006-10-11 04:21:03 -04:00
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_insert_extent:
|
|
|
|
* tries to merge requsted extent into the existing extent or
|
|
|
|
* inserts requested extent as new one into the tree,
|
|
|
|
* creating new leaf in the no-space case.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
int ext4_ext_insert_extent(handle_t *handle, struct inode *inode,
|
|
|
|
struct ext4_ext_path *path,
|
|
|
|
struct ext4_extent *newext)
|
|
|
|
{
|
|
|
|
struct ext4_extent_header * eh;
|
|
|
|
struct ext4_extent *ex, *fex;
|
|
|
|
struct ext4_extent *nearex; /* nearest extent */
|
|
|
|
struct ext4_ext_path *npath = NULL;
|
2008-01-28 23:58:27 -05:00
|
|
|
int depth, len, err;
|
|
|
|
ext4_lblk_t next;
|
2007-07-17 21:42:41 -04:00
|
|
|
unsigned uninitialized = 0;
|
2006-10-11 04:21:03 -04:00
|
|
|
|
2007-07-17 21:42:41 -04:00
|
|
|
BUG_ON(ext4_ext_get_actual_len(newext) == 0);
|
2006-10-11 04:21:03 -04:00
|
|
|
depth = ext_depth(inode);
|
|
|
|
ex = path[depth].p_ext;
|
|
|
|
BUG_ON(path[depth].p_hdr == NULL);
|
|
|
|
|
|
|
|
/* try to insert block into found extent and return */
|
|
|
|
if (ex && ext4_can_extents_be_merged(inode, ex, newext)) {
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("append %d block to %d:%d (from %llu)\n",
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_get_actual_len(newext),
|
2006-10-11 04:21:03 -04:00
|
|
|
le32_to_cpu(ex->ee_block),
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_get_actual_len(ex), ext_pblock(ex));
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_get_access(handle, inode, path + depth);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
return err;
|
2007-07-17 21:42:41 -04:00
|
|
|
|
|
|
|
/*
|
|
|
|
* ext4_can_extents_be_merged should have checked that either
|
|
|
|
* both extents are uninitialized, or both aren't. Thus we
|
|
|
|
* need to check only one of them here.
|
|
|
|
*/
|
|
|
|
if (ext4_ext_is_uninitialized(ex))
|
|
|
|
uninitialized = 1;
|
|
|
|
ex->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex)
|
|
|
|
+ ext4_ext_get_actual_len(newext));
|
|
|
|
if (uninitialized)
|
|
|
|
ext4_ext_mark_uninitialized(ex);
|
2006-10-11 04:21:03 -04:00
|
|
|
eh = path[depth].p_hdr;
|
|
|
|
nearex = ex;
|
|
|
|
goto merge;
|
|
|
|
}
|
|
|
|
|
|
|
|
repeat:
|
|
|
|
depth = ext_depth(inode);
|
|
|
|
eh = path[depth].p_hdr;
|
|
|
|
if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max))
|
|
|
|
goto has_space;
|
|
|
|
|
|
|
|
/* probably next leaf has space for us? */
|
|
|
|
fex = EXT_LAST_EXTENT(eh);
|
|
|
|
next = ext4_ext_next_leaf_block(inode, path);
|
|
|
|
if (le32_to_cpu(newext->ee_block) > le32_to_cpu(fex->ee_block)
|
|
|
|
&& next != EXT_MAX_BLOCK) {
|
|
|
|
ext_debug("next leaf block - %d\n", next);
|
|
|
|
BUG_ON(npath != NULL);
|
|
|
|
npath = ext4_ext_find_extent(inode, next, NULL);
|
|
|
|
if (IS_ERR(npath))
|
|
|
|
return PTR_ERR(npath);
|
|
|
|
BUG_ON(npath->p_depth != path->p_depth);
|
|
|
|
eh = npath[depth].p_hdr;
|
|
|
|
if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max)) {
|
|
|
|
ext_debug("next leaf isnt full(%d)\n",
|
|
|
|
le16_to_cpu(eh->eh_entries));
|
|
|
|
path = npath;
|
|
|
|
goto repeat;
|
|
|
|
}
|
|
|
|
ext_debug("next leaf has no free space(%d,%d)\n",
|
|
|
|
le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max));
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* There is no free space in the found leaf.
|
|
|
|
* We're gonna add a new leaf in the tree.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
err = ext4_ext_create_new_leaf(handle, inode, path, newext);
|
|
|
|
if (err)
|
|
|
|
goto cleanup;
|
|
|
|
depth = ext_depth(inode);
|
|
|
|
eh = path[depth].p_hdr;
|
|
|
|
|
|
|
|
has_space:
|
|
|
|
nearex = path[depth].p_ext;
|
|
|
|
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_get_access(handle, inode, path + depth);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
goto cleanup;
|
|
|
|
|
|
|
|
if (!nearex) {
|
|
|
|
/* there is no extent in this leaf, create first one */
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("first extent in the leaf: %d:%llu:%d\n",
|
2007-05-24 13:04:54 -04:00
|
|
|
le32_to_cpu(newext->ee_block),
|
|
|
|
ext_pblock(newext),
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_get_actual_len(newext));
|
2006-10-11 04:21:03 -04:00
|
|
|
path[depth].p_ext = EXT_FIRST_EXTENT(eh);
|
|
|
|
} else if (le32_to_cpu(newext->ee_block)
|
2007-05-24 13:04:54 -04:00
|
|
|
> le32_to_cpu(nearex->ee_block)) {
|
2006-10-11 04:21:03 -04:00
|
|
|
/* BUG_ON(newext->ee_block == nearex->ee_block); */
|
|
|
|
if (nearex != EXT_LAST_EXTENT(eh)) {
|
|
|
|
len = EXT_MAX_EXTENT(eh) - nearex;
|
|
|
|
len = (len - 1) * sizeof(struct ext4_extent);
|
|
|
|
len = len < 0 ? 0 : len;
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("insert %d:%llu:%d after: nearest 0x%p, "
|
2006-10-11 04:21:03 -04:00
|
|
|
"move %d from 0x%p to 0x%p\n",
|
2007-05-24 13:04:54 -04:00
|
|
|
le32_to_cpu(newext->ee_block),
|
|
|
|
ext_pblock(newext),
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_get_actual_len(newext),
|
2006-10-11 04:21:03 -04:00
|
|
|
nearex, len, nearex + 1, nearex + 2);
|
|
|
|
memmove(nearex + 2, nearex + 1, len);
|
|
|
|
}
|
|
|
|
path[depth].p_ext = nearex + 1;
|
|
|
|
} else {
|
|
|
|
BUG_ON(newext->ee_block == nearex->ee_block);
|
|
|
|
len = (EXT_MAX_EXTENT(eh) - nearex) * sizeof(struct ext4_extent);
|
|
|
|
len = len < 0 ? 0 : len;
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("insert %d:%llu:%d before: nearest 0x%p, "
|
2006-10-11 04:21:03 -04:00
|
|
|
"move %d from 0x%p to 0x%p\n",
|
|
|
|
le32_to_cpu(newext->ee_block),
|
2006-10-11 04:21:05 -04:00
|
|
|
ext_pblock(newext),
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_get_actual_len(newext),
|
2006-10-11 04:21:03 -04:00
|
|
|
nearex, len, nearex + 1, nearex + 2);
|
|
|
|
memmove(nearex + 1, nearex, len);
|
|
|
|
path[depth].p_ext = nearex;
|
|
|
|
}
|
|
|
|
|
2008-04-17 10:38:59 -04:00
|
|
|
le16_add_cpu(&eh->eh_entries, 1);
|
2006-10-11 04:21:03 -04:00
|
|
|
nearex = path[depth].p_ext;
|
|
|
|
nearex->ee_block = newext->ee_block;
|
2007-10-16 18:38:25 -04:00
|
|
|
ext4_ext_store_pblock(nearex, ext_pblock(newext));
|
2006-10-11 04:21:03 -04:00
|
|
|
nearex->ee_len = newext->ee_len;
|
|
|
|
|
|
|
|
merge:
|
|
|
|
/* try to merge extents to the right */
|
2007-07-17 21:42:38 -04:00
|
|
|
ext4_ext_try_to_merge(inode, path, nearex);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
/* try to merge extents to the left */
|
|
|
|
|
|
|
|
/* time to correct all indexes above */
|
|
|
|
err = ext4_ext_correct_indexes(handle, inode, path);
|
|
|
|
if (err)
|
|
|
|
goto cleanup;
|
|
|
|
|
|
|
|
err = ext4_ext_dirty(handle, inode, path + depth);
|
|
|
|
|
|
|
|
cleanup:
|
|
|
|
if (npath) {
|
|
|
|
ext4_ext_drop_refs(npath);
|
|
|
|
kfree(npath);
|
|
|
|
}
|
|
|
|
ext4_ext_tree_changed(inode);
|
|
|
|
ext4_ext_invalidate_cache(inode);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2006-12-06 23:41:36 -05:00
|
|
|
static void
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_ext_put_in_cache(struct inode *inode, ext4_lblk_t block,
|
2007-07-31 03:37:46 -04:00
|
|
|
__u32 len, ext4_fsblk_t start, int type)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
struct ext4_ext_cache *cex;
|
|
|
|
BUG_ON(len == 0);
|
|
|
|
cex = &EXT4_I(inode)->i_cached_extent;
|
|
|
|
cex->ec_type = type;
|
|
|
|
cex->ec_block = block;
|
|
|
|
cex->ec_len = len;
|
|
|
|
cex->ec_start = start;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_put_gap_in_cache:
|
|
|
|
* calculate boundaries of the gap that the requested block fits into
|
2006-10-11 04:21:03 -04:00
|
|
|
* and cache this gap
|
|
|
|
*/
|
2006-12-06 23:41:36 -05:00
|
|
|
static void
|
2006-10-11 04:21:03 -04:00
|
|
|
ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path,
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t block)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
int depth = ext_depth(inode);
|
2008-01-28 23:58:27 -05:00
|
|
|
unsigned long len;
|
|
|
|
ext4_lblk_t lblock;
|
2006-10-11 04:21:03 -04:00
|
|
|
struct ext4_extent *ex;
|
|
|
|
|
|
|
|
ex = path[depth].p_ext;
|
|
|
|
if (ex == NULL) {
|
|
|
|
/* there is no extent yet, so gap is [0;-] */
|
|
|
|
lblock = 0;
|
|
|
|
len = EXT_MAX_BLOCK;
|
|
|
|
ext_debug("cache gap(whole file):");
|
|
|
|
} else if (block < le32_to_cpu(ex->ee_block)) {
|
|
|
|
lblock = block;
|
|
|
|
len = le32_to_cpu(ex->ee_block) - block;
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("cache gap(before): %u [%u:%u]",
|
|
|
|
block,
|
|
|
|
le32_to_cpu(ex->ee_block),
|
|
|
|
ext4_ext_get_actual_len(ex));
|
2006-10-11 04:21:03 -04:00
|
|
|
} else if (block >= le32_to_cpu(ex->ee_block)
|
2007-07-17 21:42:41 -04:00
|
|
|
+ ext4_ext_get_actual_len(ex)) {
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t next;
|
2007-05-24 13:04:54 -04:00
|
|
|
lblock = le32_to_cpu(ex->ee_block)
|
2007-07-17 21:42:41 -04:00
|
|
|
+ ext4_ext_get_actual_len(ex);
|
2008-01-28 23:58:27 -05:00
|
|
|
|
|
|
|
next = ext4_ext_next_allocated_block(path);
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("cache gap(after): [%u:%u] %u",
|
|
|
|
le32_to_cpu(ex->ee_block),
|
|
|
|
ext4_ext_get_actual_len(ex),
|
|
|
|
block);
|
2008-01-28 23:58:27 -05:00
|
|
|
BUG_ON(next == lblock);
|
|
|
|
len = next - lblock;
|
2006-10-11 04:21:03 -04:00
|
|
|
} else {
|
|
|
|
lblock = len = 0;
|
|
|
|
BUG();
|
|
|
|
}
|
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug(" -> %u:%lu\n", lblock, len);
|
2006-10-11 04:21:03 -04:00
|
|
|
ext4_ext_put_in_cache(inode, lblock, len, 0, EXT4_EXT_CACHE_GAP);
|
|
|
|
}
|
|
|
|
|
2006-12-06 23:41:36 -05:00
|
|
|
static int
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block,
|
2006-10-11 04:21:03 -04:00
|
|
|
struct ext4_extent *ex)
|
|
|
|
{
|
|
|
|
struct ext4_ext_cache *cex;
|
|
|
|
|
|
|
|
cex = &EXT4_I(inode)->i_cached_extent;
|
|
|
|
|
|
|
|
/* has cache valid data? */
|
|
|
|
if (cex->ec_type == EXT4_EXT_CACHE_NO)
|
|
|
|
return EXT4_EXT_CACHE_NO;
|
|
|
|
|
|
|
|
BUG_ON(cex->ec_type != EXT4_EXT_CACHE_GAP &&
|
|
|
|
cex->ec_type != EXT4_EXT_CACHE_EXTENT);
|
|
|
|
if (block >= cex->ec_block && block < cex->ec_block + cex->ec_len) {
|
2007-05-24 13:04:54 -04:00
|
|
|
ex->ee_block = cpu_to_le32(cex->ec_block);
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_ext_store_pblock(ex, cex->ec_start);
|
2007-05-24 13:04:54 -04:00
|
|
|
ex->ee_len = cpu_to_le16(cex->ec_len);
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("%u cached by %u:%u:%llu\n",
|
|
|
|
block,
|
|
|
|
cex->ec_block, cex->ec_len, cex->ec_start);
|
2006-10-11 04:21:03 -04:00
|
|
|
return cex->ec_type;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* not in cache */
|
|
|
|
return EXT4_EXT_CACHE_NO;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_rm_idx:
|
|
|
|
* removes index from the index block.
|
|
|
|
* It's used in truncate case only, thus all requests are for
|
|
|
|
* last index in the block only.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
2008-01-28 23:58:27 -05:00
|
|
|
static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode,
|
2006-10-11 04:21:03 -04:00
|
|
|
struct ext4_ext_path *path)
|
|
|
|
{
|
|
|
|
struct buffer_head *bh;
|
|
|
|
int err;
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_fsblk_t leaf;
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
/* free index block */
|
|
|
|
path--;
|
2006-10-11 04:21:05 -04:00
|
|
|
leaf = idx_pblock(path->p_idx);
|
2006-10-11 04:21:03 -04:00
|
|
|
BUG_ON(path->p_hdr->eh_entries == 0);
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_get_access(handle, inode, path);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
return err;
|
2008-04-17 10:38:59 -04:00
|
|
|
le16_add_cpu(&path->p_hdr->eh_entries, -1);
|
2006-12-06 23:41:33 -05:00
|
|
|
err = ext4_ext_dirty(handle, inode, path);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
return err;
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("index is empty, remove it, free block %llu\n", leaf);
|
2006-10-11 04:21:03 -04:00
|
|
|
bh = sb_find_get_block(inode->i_sb, leaf);
|
|
|
|
ext4_forget(handle, 1, inode, bh, leaf);
|
2008-01-29 00:19:52 -05:00
|
|
|
ext4_free_blocks(handle, inode, leaf, 1, 1);
|
2006-10-11 04:21:03 -04:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_calc_credits_for_insert:
|
|
|
|
* This routine returns max. credits that the extent tree can consume.
|
2006-10-11 04:21:03 -04:00
|
|
|
* It should be OK for low-performance paths like ->writepage()
|
2006-10-11 04:21:07 -04:00
|
|
|
* To allow many writing processes to fit into a single transaction,
|
2008-01-28 23:58:26 -05:00
|
|
|
* the caller should calculate credits under i_data_sem and
|
2006-10-11 04:21:07 -04:00
|
|
|
* pass the actual path.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
2006-12-06 23:41:36 -05:00
|
|
|
int ext4_ext_calc_credits_for_insert(struct inode *inode,
|
2006-10-11 04:21:03 -04:00
|
|
|
struct ext4_ext_path *path)
|
|
|
|
{
|
|
|
|
int depth, needed;
|
|
|
|
|
|
|
|
if (path) {
|
|
|
|
/* probably there is space in leaf? */
|
|
|
|
depth = ext_depth(inode);
|
|
|
|
if (le16_to_cpu(path[depth].p_hdr->eh_entries)
|
|
|
|
< le16_to_cpu(path[depth].p_hdr->eh_max))
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* given 32-bit logical block (4294967296 blocks), max. tree
|
2006-10-11 04:21:03 -04:00
|
|
|
* can be 4 levels in depth -- 4 * 340^4 == 53453440000.
|
2006-10-11 04:21:07 -04:00
|
|
|
* Let's also add one more level for imbalance.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
depth = 5;
|
|
|
|
|
|
|
|
/* allocation of new data block(s) */
|
|
|
|
needed = 2;
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* tree can be full, so it would need to grow in depth:
|
2006-12-06 23:40:46 -05:00
|
|
|
* we need one credit to modify old root, credits for
|
|
|
|
* new root will be added in split accounting
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
2006-12-06 23:40:46 -05:00
|
|
|
needed += 1;
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* Index split can happen, we would need:
|
2006-10-11 04:21:03 -04:00
|
|
|
* allocate intermediate indexes (bitmap + group)
|
|
|
|
* + change two blocks at each level, but root (already included)
|
|
|
|
*/
|
2006-12-06 23:40:46 -05:00
|
|
|
needed += (depth * 2) + (depth * 2);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
/* any allocation modifies superblock */
|
|
|
|
needed += 1;
|
|
|
|
|
|
|
|
return needed;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ext4_remove_blocks(handle_t *handle, struct inode *inode,
|
|
|
|
struct ext4_extent *ex,
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t from, ext4_lblk_t to)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
struct buffer_head *bh;
|
2007-07-17 21:42:41 -04:00
|
|
|
unsigned short ee_len = ext4_ext_get_actual_len(ex);
|
2008-01-29 00:19:52 -05:00
|
|
|
int i, metadata = 0;
|
2006-10-11 04:21:03 -04:00
|
|
|
|
2008-01-29 00:19:52 -05:00
|
|
|
if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
|
|
|
|
metadata = 1;
|
2006-10-11 04:21:03 -04:00
|
|
|
#ifdef EXTENTS_STATS
|
|
|
|
{
|
|
|
|
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
|
|
|
|
spin_lock(&sbi->s_ext_stats_lock);
|
|
|
|
sbi->s_ext_blocks += ee_len;
|
|
|
|
sbi->s_ext_extents++;
|
|
|
|
if (ee_len < sbi->s_ext_min)
|
|
|
|
sbi->s_ext_min = ee_len;
|
|
|
|
if (ee_len > sbi->s_ext_max)
|
|
|
|
sbi->s_ext_max = ee_len;
|
|
|
|
if (ext_depth(inode) > sbi->s_depth_max)
|
|
|
|
sbi->s_depth_max = ext_depth(inode);
|
|
|
|
spin_unlock(&sbi->s_ext_stats_lock);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
if (from >= le32_to_cpu(ex->ee_block)
|
2007-07-17 21:42:41 -04:00
|
|
|
&& to == le32_to_cpu(ex->ee_block) + ee_len - 1) {
|
2006-10-11 04:21:03 -04:00
|
|
|
/* tail removal */
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t num;
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_fsblk_t start;
|
2008-01-28 23:58:27 -05:00
|
|
|
|
2007-07-17 21:42:41 -04:00
|
|
|
num = le32_to_cpu(ex->ee_block) + ee_len - from;
|
|
|
|
start = ext_pblock(ex) + ee_len - num;
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("free last %u blocks starting %llu\n", num, start);
|
2006-10-11 04:21:03 -04:00
|
|
|
for (i = 0; i < num; i++) {
|
|
|
|
bh = sb_find_get_block(inode->i_sb, start + i);
|
|
|
|
ext4_forget(handle, 0, inode, bh, start + i);
|
|
|
|
}
|
2008-01-29 00:19:52 -05:00
|
|
|
ext4_free_blocks(handle, inode, start, num, metadata);
|
2006-10-11 04:21:03 -04:00
|
|
|
} else if (from == le32_to_cpu(ex->ee_block)
|
2007-07-17 21:42:41 -04:00
|
|
|
&& to <= le32_to_cpu(ex->ee_block) + ee_len - 1) {
|
2008-01-28 23:58:27 -05:00
|
|
|
printk(KERN_INFO "strange request: removal %u-%u from %u:%u\n",
|
2007-07-17 21:42:41 -04:00
|
|
|
from, to, le32_to_cpu(ex->ee_block), ee_len);
|
2006-10-11 04:21:03 -04:00
|
|
|
} else {
|
2008-01-28 23:58:27 -05:00
|
|
|
printk(KERN_INFO "strange request: removal(2) "
|
|
|
|
"%u-%u from %u:%u\n",
|
|
|
|
from, to, le32_to_cpu(ex->ee_block), ee_len);
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
ext4_ext_rm_leaf(handle_t *handle, struct inode *inode,
|
2008-01-28 23:58:27 -05:00
|
|
|
struct ext4_ext_path *path, ext4_lblk_t start)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
int err = 0, correct_index = 0;
|
|
|
|
int depth = ext_depth(inode), credits;
|
|
|
|
struct ext4_extent_header *eh;
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t a, b, block;
|
|
|
|
unsigned num;
|
|
|
|
ext4_lblk_t ex_ee_block;
|
2006-10-11 04:21:03 -04:00
|
|
|
unsigned short ex_ee_len;
|
2007-07-17 21:42:41 -04:00
|
|
|
unsigned uninitialized = 0;
|
2006-10-11 04:21:03 -04:00
|
|
|
struct ext4_extent *ex;
|
|
|
|
|
2007-07-18 09:19:09 -04:00
|
|
|
/* the header must be checked already in ext4_ext_remove_space() */
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("truncate since %u in leaf\n", start);
|
2006-10-11 04:21:03 -04:00
|
|
|
if (!path[depth].p_hdr)
|
|
|
|
path[depth].p_hdr = ext_block_hdr(path[depth].p_bh);
|
|
|
|
eh = path[depth].p_hdr;
|
|
|
|
BUG_ON(eh == NULL);
|
|
|
|
|
|
|
|
/* find where to start removing */
|
|
|
|
ex = EXT_LAST_EXTENT(eh);
|
|
|
|
|
|
|
|
ex_ee_block = le32_to_cpu(ex->ee_block);
|
2007-07-17 21:42:41 -04:00
|
|
|
if (ext4_ext_is_uninitialized(ex))
|
|
|
|
uninitialized = 1;
|
|
|
|
ex_ee_len = ext4_ext_get_actual_len(ex);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
while (ex >= EXT_FIRST_EXTENT(eh) &&
|
|
|
|
ex_ee_block + ex_ee_len > start) {
|
|
|
|
ext_debug("remove ext %lu:%u\n", ex_ee_block, ex_ee_len);
|
|
|
|
path[depth].p_ext = ex;
|
|
|
|
|
|
|
|
a = ex_ee_block > start ? ex_ee_block : start;
|
|
|
|
b = ex_ee_block + ex_ee_len - 1 < EXT_MAX_BLOCK ?
|
|
|
|
ex_ee_block + ex_ee_len - 1 : EXT_MAX_BLOCK;
|
|
|
|
|
|
|
|
ext_debug(" border %u:%u\n", a, b);
|
|
|
|
|
|
|
|
if (a != ex_ee_block && b != ex_ee_block + ex_ee_len - 1) {
|
|
|
|
block = 0;
|
|
|
|
num = 0;
|
|
|
|
BUG();
|
|
|
|
} else if (a != ex_ee_block) {
|
|
|
|
/* remove tail of the extent */
|
|
|
|
block = ex_ee_block;
|
|
|
|
num = a - block;
|
|
|
|
} else if (b != ex_ee_block + ex_ee_len - 1) {
|
|
|
|
/* remove head of the extent */
|
|
|
|
block = a;
|
|
|
|
num = b - a;
|
|
|
|
/* there is no "make a hole" API yet */
|
|
|
|
BUG();
|
|
|
|
} else {
|
|
|
|
/* remove whole extent: excellent! */
|
|
|
|
block = ex_ee_block;
|
|
|
|
num = 0;
|
|
|
|
BUG_ON(a != ex_ee_block);
|
|
|
|
BUG_ON(b != ex_ee_block + ex_ee_len - 1);
|
|
|
|
}
|
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/* at present, extent can't cross block group: */
|
2006-10-11 04:21:03 -04:00
|
|
|
/* leaf + bitmap + group desc + sb + inode */
|
|
|
|
credits = 5;
|
|
|
|
if (ex == EXT_FIRST_EXTENT(eh)) {
|
|
|
|
correct_index = 1;
|
|
|
|
credits += (ext_depth(inode)) + 1;
|
|
|
|
}
|
|
|
|
#ifdef CONFIG_QUOTA
|
|
|
|
credits += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb);
|
|
|
|
#endif
|
|
|
|
|
2008-07-11 19:27:31 -04:00
|
|
|
err = ext4_ext_journal_restart(handle, credits);
|
|
|
|
if (err)
|
2006-10-11 04:21:03 -04:00
|
|
|
goto out;
|
|
|
|
|
|
|
|
err = ext4_ext_get_access(handle, inode, path + depth);
|
|
|
|
if (err)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
err = ext4_remove_blocks(handle, inode, ex, a, b);
|
|
|
|
if (err)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
if (num == 0) {
|
2006-10-11 04:21:07 -04:00
|
|
|
/* this extent is removed; mark slot entirely unused */
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_ext_store_pblock(ex, 0);
|
2008-04-17 10:38:59 -04:00
|
|
|
le16_add_cpu(&eh->eh_entries, -1);
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
ex->ee_block = cpu_to_le32(block);
|
|
|
|
ex->ee_len = cpu_to_le16(num);
|
2007-07-18 09:02:56 -04:00
|
|
|
/*
|
|
|
|
* Do not mark uninitialized if all the blocks in the
|
|
|
|
* extent have been removed.
|
|
|
|
*/
|
|
|
|
if (uninitialized && num)
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_mark_uninitialized(ex);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
err = ext4_ext_dirty(handle, inode, path + depth);
|
|
|
|
if (err)
|
|
|
|
goto out;
|
|
|
|
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("new extent: %u:%u:%llu\n", block, num,
|
2006-10-11 04:21:05 -04:00
|
|
|
ext_pblock(ex));
|
2006-10-11 04:21:03 -04:00
|
|
|
ex--;
|
|
|
|
ex_ee_block = le32_to_cpu(ex->ee_block);
|
2007-07-17 21:42:41 -04:00
|
|
|
ex_ee_len = ext4_ext_get_actual_len(ex);
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
if (correct_index && eh->eh_entries)
|
|
|
|
err = ext4_ext_correct_indexes(handle, inode, path);
|
|
|
|
|
|
|
|
/* if this leaf is free, then we should
|
|
|
|
* remove it from index block above */
|
|
|
|
if (err == 0 && eh->eh_entries == 0 && path[depth].p_bh != NULL)
|
|
|
|
err = ext4_ext_rm_idx(handle, inode, path + depth);
|
|
|
|
|
|
|
|
out:
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_more_to_rm:
|
|
|
|
* returns 1 if current index has to be freed (even partial)
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
2006-12-06 23:41:36 -05:00
|
|
|
static int
|
2006-10-11 04:21:03 -04:00
|
|
|
ext4_ext_more_to_rm(struct ext4_ext_path *path)
|
|
|
|
{
|
|
|
|
BUG_ON(path->p_idx == NULL);
|
|
|
|
|
|
|
|
if (path->p_idx < EXT_FIRST_INDEX(path->p_hdr))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* if truncate on deeper level happened, it wasn't partial,
|
2006-10-11 04:21:03 -04:00
|
|
|
* so we have to consider current index for truncation
|
|
|
|
*/
|
|
|
|
if (le16_to_cpu(path->p_hdr->eh_entries) == path->p_block)
|
|
|
|
return 0;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
static int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
struct super_block *sb = inode->i_sb;
|
|
|
|
int depth = ext_depth(inode);
|
|
|
|
struct ext4_ext_path *path;
|
|
|
|
handle_t *handle;
|
|
|
|
int i = 0, err = 0;
|
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("truncate since %u\n", start);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
/* probably first extent we're gonna free will be last in block */
|
|
|
|
handle = ext4_journal_start(inode, depth + 1);
|
|
|
|
if (IS_ERR(handle))
|
|
|
|
return PTR_ERR(handle);
|
|
|
|
|
|
|
|
ext4_ext_invalidate_cache(inode);
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* We start scanning from right side, freeing all the blocks
|
|
|
|
* after i_size and walking into the tree depth-wise.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
2008-04-29 22:02:02 -04:00
|
|
|
path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_NOFS);
|
2006-10-11 04:21:03 -04:00
|
|
|
if (path == NULL) {
|
|
|
|
ext4_journal_stop(handle);
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
|
|
|
path[0].p_hdr = ext_inode_hdr(inode);
|
2007-07-18 09:19:09 -04:00
|
|
|
if (ext4_ext_check_header(inode, path[0].p_hdr, depth)) {
|
2006-10-11 04:21:03 -04:00
|
|
|
err = -EIO;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
path[0].p_depth = depth;
|
|
|
|
|
|
|
|
while (i >= 0 && err == 0) {
|
|
|
|
if (i == depth) {
|
|
|
|
/* this is leaf block */
|
|
|
|
err = ext4_ext_rm_leaf(handle, inode, path, start);
|
2006-10-11 04:21:07 -04:00
|
|
|
/* root level has p_bh == NULL, brelse() eats this */
|
2006-10-11 04:21:03 -04:00
|
|
|
brelse(path[i].p_bh);
|
|
|
|
path[i].p_bh = NULL;
|
|
|
|
i--;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* this is index block */
|
|
|
|
if (!path[i].p_hdr) {
|
|
|
|
ext_debug("initialize header\n");
|
|
|
|
path[i].p_hdr = ext_block_hdr(path[i].p_bh);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!path[i].p_idx) {
|
2006-10-11 04:21:07 -04:00
|
|
|
/* this level hasn't been touched yet */
|
2006-10-11 04:21:03 -04:00
|
|
|
path[i].p_idx = EXT_LAST_INDEX(path[i].p_hdr);
|
|
|
|
path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries)+1;
|
|
|
|
ext_debug("init index ptr: hdr 0x%p, num %d\n",
|
|
|
|
path[i].p_hdr,
|
|
|
|
le16_to_cpu(path[i].p_hdr->eh_entries));
|
|
|
|
} else {
|
2006-10-11 04:21:07 -04:00
|
|
|
/* we were already here, see at next index */
|
2006-10-11 04:21:03 -04:00
|
|
|
path[i].p_idx--;
|
|
|
|
}
|
|
|
|
|
|
|
|
ext_debug("level %d - index, first 0x%p, cur 0x%p\n",
|
|
|
|
i, EXT_FIRST_INDEX(path[i].p_hdr),
|
|
|
|
path[i].p_idx);
|
|
|
|
if (ext4_ext_more_to_rm(path + i)) {
|
2007-07-18 09:19:09 -04:00
|
|
|
struct buffer_head *bh;
|
2006-10-11 04:21:03 -04:00
|
|
|
/* go to the next level */
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("move to level %d (block %llu)\n",
|
2006-10-11 04:21:05 -04:00
|
|
|
i + 1, idx_pblock(path[i].p_idx));
|
2006-10-11 04:21:03 -04:00
|
|
|
memset(path + i + 1, 0, sizeof(*path));
|
2007-07-18 09:19:09 -04:00
|
|
|
bh = sb_bread(sb, idx_pblock(path[i].p_idx));
|
|
|
|
if (!bh) {
|
2006-10-11 04:21:03 -04:00
|
|
|
/* should we reset i_size? */
|
|
|
|
err = -EIO;
|
|
|
|
break;
|
|
|
|
}
|
2007-07-18 09:19:09 -04:00
|
|
|
if (WARN_ON(i + 1 > depth)) {
|
|
|
|
err = -EIO;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (ext4_ext_check_header(inode, ext_block_hdr(bh),
|
|
|
|
depth - i - 1)) {
|
|
|
|
err = -EIO;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
path[i + 1].p_bh = bh;
|
2006-10-11 04:21:03 -04:00
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/* save actual number of indexes since this
|
|
|
|
* number is changed at the next iteration */
|
2006-10-11 04:21:03 -04:00
|
|
|
path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries);
|
|
|
|
i++;
|
|
|
|
} else {
|
2006-10-11 04:21:07 -04:00
|
|
|
/* we finished processing this index, go up */
|
2006-10-11 04:21:03 -04:00
|
|
|
if (path[i].p_hdr->eh_entries == 0 && i > 0) {
|
2006-10-11 04:21:07 -04:00
|
|
|
/* index is empty, remove it;
|
2006-10-11 04:21:03 -04:00
|
|
|
* handle must be already prepared by the
|
|
|
|
* truncatei_leaf() */
|
|
|
|
err = ext4_ext_rm_idx(handle, inode, path + i);
|
|
|
|
}
|
2006-10-11 04:21:07 -04:00
|
|
|
/* root level has p_bh == NULL, brelse() eats this */
|
2006-10-11 04:21:03 -04:00
|
|
|
brelse(path[i].p_bh);
|
|
|
|
path[i].p_bh = NULL;
|
|
|
|
i--;
|
|
|
|
ext_debug("return to level %d\n", i);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* TODO: flexible tree reduction should be here */
|
|
|
|
if (path->p_hdr->eh_entries == 0) {
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* truncate to zero freed all the tree,
|
|
|
|
* so we need to correct eh_depth
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
err = ext4_ext_get_access(handle, inode, path);
|
|
|
|
if (err == 0) {
|
|
|
|
ext_inode_hdr(inode)->eh_depth = 0;
|
|
|
|
ext_inode_hdr(inode)->eh_max =
|
|
|
|
cpu_to_le16(ext4_ext_space_root(inode));
|
|
|
|
err = ext4_ext_dirty(handle, inode, path);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
ext4_ext_tree_changed(inode);
|
|
|
|
ext4_ext_drop_refs(path);
|
|
|
|
kfree(path);
|
|
|
|
ext4_journal_stop(handle);
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* called at mount time
|
|
|
|
*/
|
|
|
|
void ext4_ext_init(struct super_block *sb)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* possible initialization would be here
|
|
|
|
*/
|
|
|
|
|
|
|
|
if (test_opt(sb, EXTENTS)) {
|
|
|
|
printk("EXT4-fs: file extents enabled");
|
2007-02-17 13:20:16 -05:00
|
|
|
#ifdef AGGRESSIVE_TEST
|
|
|
|
printk(", aggressive tests");
|
2006-10-11 04:21:03 -04:00
|
|
|
#endif
|
|
|
|
#ifdef CHECK_BINSEARCH
|
|
|
|
printk(", check binsearch");
|
|
|
|
#endif
|
|
|
|
#ifdef EXTENTS_STATS
|
|
|
|
printk(", stats");
|
|
|
|
#endif
|
|
|
|
printk("\n");
|
|
|
|
#ifdef EXTENTS_STATS
|
|
|
|
spin_lock_init(&EXT4_SB(sb)->s_ext_stats_lock);
|
|
|
|
EXT4_SB(sb)->s_ext_min = 1 << 30;
|
|
|
|
EXT4_SB(sb)->s_ext_max = 0;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* called at umount time
|
|
|
|
*/
|
|
|
|
void ext4_ext_release(struct super_block *sb)
|
|
|
|
{
|
|
|
|
if (!test_opt(sb, EXTENTS))
|
|
|
|
return;
|
|
|
|
|
|
|
|
#ifdef EXTENTS_STATS
|
|
|
|
if (EXT4_SB(sb)->s_ext_blocks && EXT4_SB(sb)->s_ext_extents) {
|
|
|
|
struct ext4_sb_info *sbi = EXT4_SB(sb);
|
|
|
|
printk(KERN_ERR "EXT4-fs: %lu blocks in %lu extents (%lu ave)\n",
|
|
|
|
sbi->s_ext_blocks, sbi->s_ext_extents,
|
|
|
|
sbi->s_ext_blocks / sbi->s_ext_extents);
|
|
|
|
printk(KERN_ERR "EXT4-fs: extents: %lu min, %lu max, max depth %lu\n",
|
|
|
|
sbi->s_ext_min, sbi->s_ext_max, sbi->s_depth_max);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2008-04-29 08:11:12 -04:00
|
|
|
static void bi_complete(struct bio *bio, int error)
|
|
|
|
{
|
|
|
|
complete((struct completion *)bio->bi_private);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* FIXME!! we need to try to merge to left or right after zero-out */
|
|
|
|
static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex)
|
|
|
|
{
|
|
|
|
int ret = -EIO;
|
|
|
|
struct bio *bio;
|
|
|
|
int blkbits, blocksize;
|
|
|
|
sector_t ee_pblock;
|
|
|
|
struct completion event;
|
|
|
|
unsigned int ee_len, len, done, offset;
|
|
|
|
|
|
|
|
|
|
|
|
blkbits = inode->i_blkbits;
|
|
|
|
blocksize = inode->i_sb->s_blocksize;
|
|
|
|
ee_len = ext4_ext_get_actual_len(ex);
|
|
|
|
ee_pblock = ext_pblock(ex);
|
|
|
|
|
|
|
|
/* convert ee_pblock to 512 byte sectors */
|
|
|
|
ee_pblock = ee_pblock << (blkbits - 9);
|
|
|
|
|
|
|
|
while (ee_len > 0) {
|
|
|
|
|
|
|
|
if (ee_len > BIO_MAX_PAGES)
|
|
|
|
len = BIO_MAX_PAGES;
|
|
|
|
else
|
|
|
|
len = ee_len;
|
|
|
|
|
|
|
|
bio = bio_alloc(GFP_NOIO, len);
|
|
|
|
if (!bio)
|
|
|
|
return -ENOMEM;
|
|
|
|
bio->bi_sector = ee_pblock;
|
|
|
|
bio->bi_bdev = inode->i_sb->s_bdev;
|
|
|
|
|
|
|
|
done = 0;
|
|
|
|
offset = 0;
|
|
|
|
while (done < len) {
|
|
|
|
ret = bio_add_page(bio, ZERO_PAGE(0),
|
|
|
|
blocksize, offset);
|
|
|
|
if (ret != blocksize) {
|
|
|
|
/*
|
|
|
|
* We can't add any more pages because of
|
|
|
|
* hardware limitations. Start a new bio.
|
|
|
|
*/
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
done++;
|
|
|
|
offset += blocksize;
|
|
|
|
if (offset >= PAGE_CACHE_SIZE)
|
|
|
|
offset = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
init_completion(&event);
|
|
|
|
bio->bi_private = &event;
|
|
|
|
bio->bi_end_io = bi_complete;
|
|
|
|
submit_bio(WRITE, bio);
|
|
|
|
wait_for_completion(&event);
|
|
|
|
|
|
|
|
if (test_bit(BIO_UPTODATE, &bio->bi_flags))
|
|
|
|
ret = 0;
|
|
|
|
else {
|
|
|
|
ret = -EIO;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
bio_put(bio);
|
|
|
|
ee_len -= done;
|
|
|
|
ee_pblock += done << (blkbits - 9);
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2008-04-17 10:38:59 -04:00
|
|
|
#define EXT4_EXT_ZERO_LEN 7
|
|
|
|
|
2007-07-17 21:42:38 -04:00
|
|
|
/*
|
|
|
|
* This function is called by ext4_ext_get_blocks() if someone tries to write
|
|
|
|
* to an uninitialized extent. It may result in splitting the uninitialized
|
|
|
|
* extent into multiple extents (upto three - one initialized and two
|
|
|
|
* uninitialized).
|
|
|
|
* There are three possibilities:
|
|
|
|
* a> There is no split required: Entire extent should be initialized
|
|
|
|
* b> Splits in two extents: Write is happening at either end of the extent
|
|
|
|
* c> Splits in three extents: Somone is writing in middle of the extent
|
|
|
|
*/
|
2008-01-28 23:58:27 -05:00
|
|
|
static int ext4_ext_convert_to_initialized(handle_t *handle,
|
|
|
|
struct inode *inode,
|
|
|
|
struct ext4_ext_path *path,
|
|
|
|
ext4_lblk_t iblock,
|
|
|
|
unsigned long max_blocks)
|
2007-07-17 21:42:38 -04:00
|
|
|
{
|
2008-04-17 10:38:59 -04:00
|
|
|
struct ext4_extent *ex, newex, orig_ex;
|
2007-07-17 21:42:38 -04:00
|
|
|
struct ext4_extent *ex1 = NULL;
|
|
|
|
struct ext4_extent *ex2 = NULL;
|
|
|
|
struct ext4_extent *ex3 = NULL;
|
|
|
|
struct ext4_extent_header *eh;
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t ee_block;
|
|
|
|
unsigned int allocated, ee_len, depth;
|
2007-07-17 21:42:38 -04:00
|
|
|
ext4_fsblk_t newblock;
|
|
|
|
int err = 0;
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
depth = ext_depth(inode);
|
|
|
|
eh = path[depth].p_hdr;
|
|
|
|
ex = path[depth].p_ext;
|
|
|
|
ee_block = le32_to_cpu(ex->ee_block);
|
|
|
|
ee_len = ext4_ext_get_actual_len(ex);
|
|
|
|
allocated = ee_len - (iblock - ee_block);
|
|
|
|
newblock = iblock - ee_block + ext_pblock(ex);
|
|
|
|
ex2 = ex;
|
2008-04-17 10:38:59 -04:00
|
|
|
orig_ex.ee_block = ex->ee_block;
|
|
|
|
orig_ex.ee_len = cpu_to_le16(ee_len);
|
|
|
|
ext4_ext_store_pblock(&orig_ex, ext_pblock(ex));
|
2007-07-17 21:42:38 -04:00
|
|
|
|
2008-02-22 06:17:31 -05:00
|
|
|
err = ext4_ext_get_access(handle, inode, path + depth);
|
|
|
|
if (err)
|
|
|
|
goto out;
|
2008-04-17 10:38:59 -04:00
|
|
|
/* If extent has less than 2*EXT4_EXT_ZERO_LEN zerout directly */
|
|
|
|
if (ee_len <= 2*EXT4_EXT_ZERO_LEN) {
|
|
|
|
err = ext4_ext_zeroout(inode, &orig_ex);
|
|
|
|
if (err)
|
|
|
|
goto fix_extent_len;
|
|
|
|
/* update the extent length and mark as initialized */
|
|
|
|
ex->ee_block = orig_ex.ee_block;
|
|
|
|
ex->ee_len = orig_ex.ee_len;
|
|
|
|
ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
|
|
|
|
ext4_ext_dirty(handle, inode, path + depth);
|
2008-04-29 22:03:59 -04:00
|
|
|
/* zeroed the full extent */
|
|
|
|
return allocated;
|
2008-04-17 10:38:59 -04:00
|
|
|
}
|
2008-02-22 06:17:31 -05:00
|
|
|
|
2007-07-17 21:42:38 -04:00
|
|
|
/* ex1: ee_block to iblock - 1 : uninitialized */
|
|
|
|
if (iblock > ee_block) {
|
|
|
|
ex1 = ex;
|
|
|
|
ex1->ee_len = cpu_to_le16(iblock - ee_block);
|
|
|
|
ext4_ext_mark_uninitialized(ex1);
|
|
|
|
ex2 = &newex;
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* for sanity, update the length of the ex2 extent before
|
|
|
|
* we insert ex3, if ex1 is NULL. This is to avoid temporary
|
|
|
|
* overlap of blocks.
|
|
|
|
*/
|
|
|
|
if (!ex1 && allocated > max_blocks)
|
|
|
|
ex2->ee_len = cpu_to_le16(max_blocks);
|
|
|
|
/* ex3: to ee_block + ee_len : uninitialised */
|
|
|
|
if (allocated > max_blocks) {
|
|
|
|
unsigned int newdepth;
|
2008-04-17 10:38:59 -04:00
|
|
|
/* If extent has less than EXT4_EXT_ZERO_LEN zerout directly */
|
|
|
|
if (allocated <= EXT4_EXT_ZERO_LEN) {
|
|
|
|
/* Mark first half uninitialized.
|
|
|
|
* Mark second half initialized and zero out the
|
|
|
|
* initialized extent
|
|
|
|
*/
|
|
|
|
ex->ee_block = orig_ex.ee_block;
|
|
|
|
ex->ee_len = cpu_to_le16(ee_len - allocated);
|
|
|
|
ext4_ext_mark_uninitialized(ex);
|
|
|
|
ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
|
|
|
|
ext4_ext_dirty(handle, inode, path + depth);
|
|
|
|
|
|
|
|
ex3 = &newex;
|
|
|
|
ex3->ee_block = cpu_to_le32(iblock);
|
|
|
|
ext4_ext_store_pblock(ex3, newblock);
|
|
|
|
ex3->ee_len = cpu_to_le16(allocated);
|
|
|
|
err = ext4_ext_insert_extent(handle, inode, path, ex3);
|
|
|
|
if (err == -ENOSPC) {
|
|
|
|
err = ext4_ext_zeroout(inode, &orig_ex);
|
|
|
|
if (err)
|
|
|
|
goto fix_extent_len;
|
|
|
|
ex->ee_block = orig_ex.ee_block;
|
|
|
|
ex->ee_len = orig_ex.ee_len;
|
|
|
|
ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
|
|
|
|
ext4_ext_dirty(handle, inode, path + depth);
|
2008-04-29 22:03:59 -04:00
|
|
|
/* zeroed the full extent */
|
|
|
|
return allocated;
|
2008-04-17 10:38:59 -04:00
|
|
|
|
|
|
|
} else if (err)
|
|
|
|
goto fix_extent_len;
|
|
|
|
|
2008-04-29 22:03:59 -04:00
|
|
|
/*
|
|
|
|
* We need to zero out the second half because
|
|
|
|
* an fallocate request can update file size and
|
|
|
|
* converting the second half to initialized extent
|
|
|
|
* implies that we can leak some junk data to user
|
|
|
|
* space.
|
|
|
|
*/
|
|
|
|
err = ext4_ext_zeroout(inode, ex3);
|
|
|
|
if (err) {
|
|
|
|
/*
|
|
|
|
* We should actually mark the
|
|
|
|
* second half as uninit and return error
|
|
|
|
* Insert would have changed the extent
|
|
|
|
*/
|
|
|
|
depth = ext_depth(inode);
|
|
|
|
ext4_ext_drop_refs(path);
|
|
|
|
path = ext4_ext_find_extent(inode,
|
|
|
|
iblock, path);
|
|
|
|
if (IS_ERR(path)) {
|
|
|
|
err = PTR_ERR(path);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
ex = path[depth].p_ext;
|
|
|
|
err = ext4_ext_get_access(handle, inode,
|
|
|
|
path + depth);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
ext4_ext_mark_uninitialized(ex);
|
|
|
|
ext4_ext_dirty(handle, inode, path + depth);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* zeroed the second half */
|
2008-04-17 10:38:59 -04:00
|
|
|
return allocated;
|
|
|
|
}
|
2007-07-17 21:42:38 -04:00
|
|
|
ex3 = &newex;
|
|
|
|
ex3->ee_block = cpu_to_le32(iblock + max_blocks);
|
|
|
|
ext4_ext_store_pblock(ex3, newblock + max_blocks);
|
|
|
|
ex3->ee_len = cpu_to_le16(allocated - max_blocks);
|
|
|
|
ext4_ext_mark_uninitialized(ex3);
|
|
|
|
err = ext4_ext_insert_extent(handle, inode, path, ex3);
|
2008-04-29 08:11:12 -04:00
|
|
|
if (err == -ENOSPC) {
|
|
|
|
err = ext4_ext_zeroout(inode, &orig_ex);
|
|
|
|
if (err)
|
|
|
|
goto fix_extent_len;
|
|
|
|
/* update the extent length and mark as initialized */
|
2008-04-17 10:38:59 -04:00
|
|
|
ex->ee_block = orig_ex.ee_block;
|
|
|
|
ex->ee_len = orig_ex.ee_len;
|
|
|
|
ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
|
|
|
|
ext4_ext_dirty(handle, inode, path + depth);
|
2008-04-29 22:03:59 -04:00
|
|
|
/* zeroed the full extent */
|
|
|
|
return allocated;
|
2008-04-29 08:11:12 -04:00
|
|
|
|
|
|
|
} else if (err)
|
|
|
|
goto fix_extent_len;
|
2007-07-17 21:42:38 -04:00
|
|
|
/*
|
|
|
|
* The depth, and hence eh & ex might change
|
|
|
|
* as part of the insert above.
|
|
|
|
*/
|
|
|
|
newdepth = ext_depth(inode);
|
2008-04-17 10:38:59 -04:00
|
|
|
/*
|
|
|
|
* update the extent length after successfull insert of the
|
|
|
|
* split extent
|
|
|
|
*/
|
|
|
|
orig_ex.ee_len = cpu_to_le16(ee_len -
|
|
|
|
ext4_ext_get_actual_len(ex3));
|
2007-07-17 21:42:38 -04:00
|
|
|
if (newdepth != depth) {
|
|
|
|
depth = newdepth;
|
2008-02-25 16:54:37 -05:00
|
|
|
ext4_ext_drop_refs(path);
|
|
|
|
path = ext4_ext_find_extent(inode, iblock, path);
|
2007-07-17 21:42:38 -04:00
|
|
|
if (IS_ERR(path)) {
|
|
|
|
err = PTR_ERR(path);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
eh = path[depth].p_hdr;
|
|
|
|
ex = path[depth].p_ext;
|
|
|
|
if (ex2 != &newex)
|
|
|
|
ex2 = ex;
|
2008-02-22 06:17:31 -05:00
|
|
|
|
|
|
|
err = ext4_ext_get_access(handle, inode, path + depth);
|
|
|
|
if (err)
|
|
|
|
goto out;
|
2007-07-17 21:42:38 -04:00
|
|
|
}
|
|
|
|
allocated = max_blocks;
|
2008-04-17 10:38:59 -04:00
|
|
|
|
|
|
|
/* If extent has less than EXT4_EXT_ZERO_LEN and we are trying
|
|
|
|
* to insert a extent in the middle zerout directly
|
|
|
|
* otherwise give the extent a chance to merge to left
|
|
|
|
*/
|
|
|
|
if (le16_to_cpu(orig_ex.ee_len) <= EXT4_EXT_ZERO_LEN &&
|
|
|
|
iblock != ee_block) {
|
|
|
|
err = ext4_ext_zeroout(inode, &orig_ex);
|
|
|
|
if (err)
|
|
|
|
goto fix_extent_len;
|
|
|
|
/* update the extent length and mark as initialized */
|
|
|
|
ex->ee_block = orig_ex.ee_block;
|
|
|
|
ex->ee_len = orig_ex.ee_len;
|
|
|
|
ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
|
|
|
|
ext4_ext_dirty(handle, inode, path + depth);
|
2008-04-29 22:03:59 -04:00
|
|
|
/* zero out the first half */
|
|
|
|
return allocated;
|
2008-04-17 10:38:59 -04:00
|
|
|
}
|
2007-07-17 21:42:38 -04:00
|
|
|
}
|
|
|
|
/*
|
|
|
|
* If there was a change of depth as part of the
|
|
|
|
* insertion of ex3 above, we need to update the length
|
|
|
|
* of the ex1 extent again here
|
|
|
|
*/
|
|
|
|
if (ex1 && ex1 != ex) {
|
|
|
|
ex1 = ex;
|
|
|
|
ex1->ee_len = cpu_to_le16(iblock - ee_block);
|
|
|
|
ext4_ext_mark_uninitialized(ex1);
|
|
|
|
ex2 = &newex;
|
|
|
|
}
|
|
|
|
/* ex2: iblock to iblock + maxblocks-1 : initialised */
|
|
|
|
ex2->ee_block = cpu_to_le32(iblock);
|
|
|
|
ext4_ext_store_pblock(ex2, newblock);
|
|
|
|
ex2->ee_len = cpu_to_le16(allocated);
|
|
|
|
if (ex2 != ex)
|
|
|
|
goto insert;
|
|
|
|
/*
|
|
|
|
* New (initialized) extent starts from the first block
|
|
|
|
* in the current extent. i.e., ex2 == ex
|
|
|
|
* We have to see if it can be merged with the extent
|
|
|
|
* on the left.
|
|
|
|
*/
|
|
|
|
if (ex2 > EXT_FIRST_EXTENT(eh)) {
|
|
|
|
/*
|
|
|
|
* To merge left, pass "ex2 - 1" to try_to_merge(),
|
|
|
|
* since it merges towards right _only_.
|
|
|
|
*/
|
|
|
|
ret = ext4_ext_try_to_merge(inode, path, ex2 - 1);
|
|
|
|
if (ret) {
|
|
|
|
err = ext4_ext_correct_indexes(handle, inode, path);
|
|
|
|
if (err)
|
|
|
|
goto out;
|
|
|
|
depth = ext_depth(inode);
|
|
|
|
ex2--;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* Try to Merge towards right. This might be required
|
|
|
|
* only when the whole extent is being written to.
|
|
|
|
* i.e. ex2 == ex and ex3 == NULL.
|
|
|
|
*/
|
|
|
|
if (!ex3) {
|
|
|
|
ret = ext4_ext_try_to_merge(inode, path, ex2);
|
|
|
|
if (ret) {
|
|
|
|
err = ext4_ext_correct_indexes(handle, inode, path);
|
|
|
|
if (err)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
/* Mark modified extent as dirty */
|
|
|
|
err = ext4_ext_dirty(handle, inode, path + depth);
|
|
|
|
goto out;
|
|
|
|
insert:
|
|
|
|
err = ext4_ext_insert_extent(handle, inode, path, &newex);
|
2008-04-29 08:11:12 -04:00
|
|
|
if (err == -ENOSPC) {
|
|
|
|
err = ext4_ext_zeroout(inode, &orig_ex);
|
|
|
|
if (err)
|
|
|
|
goto fix_extent_len;
|
|
|
|
/* update the extent length and mark as initialized */
|
2008-04-17 10:38:59 -04:00
|
|
|
ex->ee_block = orig_ex.ee_block;
|
|
|
|
ex->ee_len = orig_ex.ee_len;
|
|
|
|
ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
|
|
|
|
ext4_ext_dirty(handle, inode, path + depth);
|
2008-04-29 22:03:59 -04:00
|
|
|
/* zero out the first half */
|
|
|
|
return allocated;
|
2008-04-29 08:11:12 -04:00
|
|
|
} else if (err)
|
|
|
|
goto fix_extent_len;
|
2007-07-17 21:42:38 -04:00
|
|
|
out:
|
|
|
|
return err ? err : allocated;
|
2008-04-29 08:11:12 -04:00
|
|
|
|
|
|
|
fix_extent_len:
|
|
|
|
ex->ee_block = orig_ex.ee_block;
|
|
|
|
ex->ee_len = orig_ex.ee_len;
|
|
|
|
ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
|
|
|
|
ext4_ext_mark_uninitialized(ex);
|
|
|
|
ext4_ext_dirty(handle, inode, path + depth);
|
|
|
|
return err;
|
2007-07-17 21:42:38 -04:00
|
|
|
}
|
|
|
|
|
2008-01-28 23:58:27 -05:00
|
|
|
/*
|
2008-02-25 15:29:55 -05:00
|
|
|
* Block allocation/map/preallocation routine for extents based files
|
|
|
|
*
|
|
|
|
*
|
2008-01-28 23:58:27 -05:00
|
|
|
* Need to be called with
|
2008-01-28 23:58:26 -05:00
|
|
|
* down_read(&EXT4_I(inode)->i_data_sem) if not allocating file system block
|
|
|
|
* (ie, create is zero). Otherwise down_write(&EXT4_I(inode)->i_data_sem)
|
2008-02-25 15:29:55 -05:00
|
|
|
*
|
|
|
|
* return > 0, number of of blocks already mapped/allocated
|
|
|
|
* if create == 0 and these are pre-allocated blocks
|
|
|
|
* buffer head is unmapped
|
|
|
|
* otherwise blocks are mapped
|
|
|
|
*
|
|
|
|
* return = 0, if plain look up failed (blocks have not been allocated)
|
|
|
|
* buffer head is unmapped
|
|
|
|
*
|
|
|
|
* return < 0, error case.
|
2008-01-28 23:58:27 -05:00
|
|
|
*/
|
2006-10-11 04:21:05 -04:00
|
|
|
int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t iblock,
|
2006-10-11 04:21:03 -04:00
|
|
|
unsigned long max_blocks, struct buffer_head *bh_result,
|
|
|
|
int create, int extend_disksize)
|
|
|
|
{
|
|
|
|
struct ext4_ext_path *path = NULL;
|
2007-07-17 21:42:38 -04:00
|
|
|
struct ext4_extent_header *eh;
|
2006-10-11 04:21:03 -04:00
|
|
|
struct ext4_extent newex, *ex;
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_fsblk_t goal, newblock;
|
2007-07-17 21:42:38 -04:00
|
|
|
int err = 0, depth, ret;
|
2006-10-11 04:21:03 -04:00
|
|
|
unsigned long allocated = 0;
|
2008-01-29 00:19:52 -05:00
|
|
|
struct ext4_allocation_request ar;
|
2008-07-11 19:27:31 -04:00
|
|
|
loff_t disksize;
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
__clear_bit(BH_New, &bh_result->b_state);
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("blocks %u/%lu requested for inode %u\n",
|
|
|
|
iblock, max_blocks, inode->i_ino);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
/* check in cache */
|
2006-12-06 23:41:33 -05:00
|
|
|
goal = ext4_ext_in_cache(inode, iblock, &newex);
|
|
|
|
if (goal) {
|
2006-10-11 04:21:03 -04:00
|
|
|
if (goal == EXT4_EXT_CACHE_GAP) {
|
|
|
|
if (!create) {
|
2007-07-17 21:42:38 -04:00
|
|
|
/*
|
|
|
|
* block isn't allocated yet and
|
|
|
|
* user doesn't want to allocate it
|
|
|
|
*/
|
2006-10-11 04:21:03 -04:00
|
|
|
goto out2;
|
|
|
|
}
|
|
|
|
/* we should allocate requested block */
|
|
|
|
} else if (goal == EXT4_EXT_CACHE_EXTENT) {
|
|
|
|
/* block is already allocated */
|
2007-05-24 13:04:54 -04:00
|
|
|
newblock = iblock
|
|
|
|
- le32_to_cpu(newex.ee_block)
|
|
|
|
+ ext_pblock(&newex);
|
2006-10-11 04:21:07 -04:00
|
|
|
/* number of remaining blocks in the extent */
|
2008-01-28 23:58:27 -05:00
|
|
|
allocated = ext4_ext_get_actual_len(&newex) -
|
2006-10-11 04:21:03 -04:00
|
|
|
(iblock - le32_to_cpu(newex.ee_block));
|
|
|
|
goto out;
|
|
|
|
} else {
|
|
|
|
BUG();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* find extent for this block */
|
|
|
|
path = ext4_ext_find_extent(inode, iblock, NULL);
|
|
|
|
if (IS_ERR(path)) {
|
|
|
|
err = PTR_ERR(path);
|
|
|
|
path = NULL;
|
|
|
|
goto out2;
|
|
|
|
}
|
|
|
|
|
|
|
|
depth = ext_depth(inode);
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* consistent leaf must not be empty;
|
|
|
|
* this situation is possible, though, _during_ tree modification;
|
2006-10-11 04:21:03 -04:00
|
|
|
* this is why assert can't be put in ext4_ext_find_extent()
|
|
|
|
*/
|
|
|
|
BUG_ON(path[depth].p_ext == NULL && depth != 0);
|
2007-07-17 21:42:38 -04:00
|
|
|
eh = path[depth].p_hdr;
|
2006-10-11 04:21:03 -04:00
|
|
|
|
2006-12-06 23:41:33 -05:00
|
|
|
ex = path[depth].p_ext;
|
|
|
|
if (ex) {
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t ee_block = le32_to_cpu(ex->ee_block);
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_fsblk_t ee_start = ext_pblock(ex);
|
2007-07-17 21:42:41 -04:00
|
|
|
unsigned short ee_len;
|
2006-10-11 04:21:06 -04:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Uninitialized extents are treated as holes, except that
|
2007-07-17 21:42:38 -04:00
|
|
|
* we split out initialized portions during a write.
|
2006-10-11 04:21:06 -04:00
|
|
|
*/
|
2007-07-17 21:42:41 -04:00
|
|
|
ee_len = ext4_ext_get_actual_len(ex);
|
2006-10-11 04:21:07 -04:00
|
|
|
/* if found extent covers block, simply return it */
|
2007-05-24 13:04:54 -04:00
|
|
|
if (iblock >= ee_block && iblock < ee_block + ee_len) {
|
2006-10-11 04:21:03 -04:00
|
|
|
newblock = iblock - ee_block + ee_start;
|
2006-10-11 04:21:07 -04:00
|
|
|
/* number of remaining blocks in the extent */
|
2006-10-11 04:21:03 -04:00
|
|
|
allocated = ee_len - (iblock - ee_block);
|
2008-01-28 23:58:27 -05:00
|
|
|
ext_debug("%u fit into %lu:%d -> %llu\n", iblock,
|
2006-10-11 04:21:03 -04:00
|
|
|
ee_block, ee_len, newblock);
|
2007-07-17 21:42:38 -04:00
|
|
|
|
2007-07-17 21:42:41 -04:00
|
|
|
/* Do not put uninitialized extent in the cache */
|
2007-07-17 21:42:38 -04:00
|
|
|
if (!ext4_ext_is_uninitialized(ex)) {
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_ext_put_in_cache(inode, ee_block,
|
|
|
|
ee_len, ee_start,
|
|
|
|
EXT4_EXT_CACHE_EXTENT);
|
2007-07-17 21:42:38 -04:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
if (create == EXT4_CREATE_UNINITIALIZED_EXT)
|
|
|
|
goto out;
|
2008-04-29 08:11:12 -04:00
|
|
|
if (!create) {
|
|
|
|
/*
|
|
|
|
* We have blocks reserved already. We
|
|
|
|
* return allocated blocks so that delalloc
|
|
|
|
* won't do block reservation for us. But
|
|
|
|
* the buffer head will be unmapped so that
|
|
|
|
* a read from the block returns 0s.
|
|
|
|
*/
|
|
|
|
if (allocated > max_blocks)
|
|
|
|
allocated = max_blocks;
|
2008-07-11 19:27:31 -04:00
|
|
|
set_buffer_unwritten(bh_result);
|
2007-07-17 21:42:38 -04:00
|
|
|
goto out2;
|
2008-04-29 08:11:12 -04:00
|
|
|
}
|
2007-07-17 21:42:38 -04:00
|
|
|
|
|
|
|
ret = ext4_ext_convert_to_initialized(handle, inode,
|
|
|
|
path, iblock,
|
|
|
|
max_blocks);
|
ext4: fix uniniatilized extent splitting error
Fix bug reported by Dmitry Monakhov caused by lost error code
Testcase:
blksize = 0x1000;
fd = open(argv[1], O_RDWR|O_CREAT, 0700);
unsigned long long sz = 0x10000000UL;
/* allocating big blocks chunk */
syscall(__NR_fallocate, fd, 0, 0UL, sz)
/* grab all other available filesystem space */
tfd = open("tmp", O_RDWR|O_CREAT|O_DIRECT, 0700);
while( write(tfd, buf, 4096) > 0); /* loop untill ENOSPC */
fsync(fd); /* just in case */
while (pos < sz) {
/* each seek+ write operation result in splits uninitialized extent
in three extents. Splitting may result in new extent allocation
which probably will fail because of ENOSPC*/
lseek(fd, blksize*2 -1, SEEK_CUR);
if ((ret = write(fd, 'a', 1)) != 1)
exit(1);
pos += blksize * 2;
}
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-01-28 23:58:27 -05:00
|
|
|
if (ret <= 0) {
|
|
|
|
err = ret;
|
2007-07-17 21:42:38 -04:00
|
|
|
goto out2;
|
ext4: fix uniniatilized extent splitting error
Fix bug reported by Dmitry Monakhov caused by lost error code
Testcase:
blksize = 0x1000;
fd = open(argv[1], O_RDWR|O_CREAT, 0700);
unsigned long long sz = 0x10000000UL;
/* allocating big blocks chunk */
syscall(__NR_fallocate, fd, 0, 0UL, sz)
/* grab all other available filesystem space */
tfd = open("tmp", O_RDWR|O_CREAT|O_DIRECT, 0700);
while( write(tfd, buf, 4096) > 0); /* loop untill ENOSPC */
fsync(fd); /* just in case */
while (pos < sz) {
/* each seek+ write operation result in splits uninitialized extent
in three extents. Splitting may result in new extent allocation
which probably will fail because of ENOSPC*/
lseek(fd, blksize*2 -1, SEEK_CUR);
if ((ret = write(fd, 'a', 1)) != 1)
exit(1);
pos += blksize * 2;
}
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-01-28 23:58:27 -05:00
|
|
|
} else
|
2007-07-17 21:42:38 -04:00
|
|
|
allocated = ret;
|
|
|
|
goto outnew;
|
2006-10-11 04:21:03 -04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* requested block isn't allocated yet;
|
2006-10-11 04:21:03 -04:00
|
|
|
* we couldn't try to create block if create flag is zero
|
|
|
|
*/
|
|
|
|
if (!create) {
|
2007-07-17 21:42:38 -04:00
|
|
|
/*
|
|
|
|
* put just found gap into cache to speed up
|
|
|
|
* subsequent requests
|
|
|
|
*/
|
2006-10-11 04:21:03 -04:00
|
|
|
ext4_ext_put_gap_in_cache(inode, path, iblock);
|
|
|
|
goto out2;
|
|
|
|
}
|
|
|
|
/*
|
2006-10-11 04:21:24 -04:00
|
|
|
* Okay, we need to do block allocation. Lazily initialize the block
|
|
|
|
* allocation info here if necessary.
|
|
|
|
*/
|
2006-10-11 04:21:03 -04:00
|
|
|
if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info))
|
|
|
|
ext4_init_block_alloc_info(inode);
|
|
|
|
|
2008-01-29 00:19:52 -05:00
|
|
|
/* find neighbour allocated blocks */
|
|
|
|
ar.lleft = iblock;
|
|
|
|
err = ext4_ext_search_left(inode, path, &ar.lleft, &ar.pleft);
|
|
|
|
if (err)
|
|
|
|
goto out2;
|
|
|
|
ar.lright = iblock;
|
|
|
|
err = ext4_ext_search_right(inode, path, &ar.lright, &ar.pright);
|
|
|
|
if (err)
|
|
|
|
goto out2;
|
2007-05-24 13:04:13 -04:00
|
|
|
|
2007-07-18 09:02:56 -04:00
|
|
|
/*
|
|
|
|
* See if request is beyond maximum number of blocks we can have in
|
|
|
|
* a single extent. For an initialized extent this limit is
|
|
|
|
* EXT_INIT_MAX_LEN and for an uninitialized extent this limit is
|
|
|
|
* EXT_UNINIT_MAX_LEN.
|
|
|
|
*/
|
|
|
|
if (max_blocks > EXT_INIT_MAX_LEN &&
|
|
|
|
create != EXT4_CREATE_UNINITIALIZED_EXT)
|
|
|
|
max_blocks = EXT_INIT_MAX_LEN;
|
|
|
|
else if (max_blocks > EXT_UNINIT_MAX_LEN &&
|
|
|
|
create == EXT4_CREATE_UNINITIALIZED_EXT)
|
|
|
|
max_blocks = EXT_UNINIT_MAX_LEN;
|
|
|
|
|
2007-05-24 13:04:13 -04:00
|
|
|
/* Check if we can really insert (iblock)::(iblock+max_blocks) extent */
|
|
|
|
newex.ee_block = cpu_to_le32(iblock);
|
|
|
|
newex.ee_len = cpu_to_le16(max_blocks);
|
|
|
|
err = ext4_ext_check_overlap(inode, &newex, path);
|
|
|
|
if (err)
|
2008-01-28 23:58:27 -05:00
|
|
|
allocated = ext4_ext_get_actual_len(&newex);
|
2007-05-24 13:04:13 -04:00
|
|
|
else
|
|
|
|
allocated = max_blocks;
|
2008-01-29 00:19:52 -05:00
|
|
|
|
|
|
|
/* allocate new block */
|
|
|
|
ar.inode = inode;
|
|
|
|
ar.goal = ext4_ext_find_goal(inode, path, iblock);
|
|
|
|
ar.logical = iblock;
|
|
|
|
ar.len = allocated;
|
|
|
|
if (S_ISREG(inode->i_mode))
|
|
|
|
ar.flags = EXT4_MB_HINT_DATA;
|
|
|
|
else
|
|
|
|
/* disable in-core preallocation for non-regular files */
|
|
|
|
ar.flags = 0;
|
|
|
|
newblock = ext4_mb_new_blocks(handle, &ar, &err);
|
2006-10-11 04:21:03 -04:00
|
|
|
if (!newblock)
|
|
|
|
goto out2;
|
2006-10-11 04:21:11 -04:00
|
|
|
ext_debug("allocate new block: goal %llu, found %llu/%lu\n",
|
2006-10-11 04:21:03 -04:00
|
|
|
goal, newblock, allocated);
|
|
|
|
|
|
|
|
/* try to insert new extent into found leaf and return */
|
2006-10-11 04:21:05 -04:00
|
|
|
ext4_ext_store_pblock(&newex, newblock);
|
2008-01-29 00:19:52 -05:00
|
|
|
newex.ee_len = cpu_to_le16(ar.len);
|
2007-07-17 21:42:41 -04:00
|
|
|
if (create == EXT4_CREATE_UNINITIALIZED_EXT) /* Mark uninitialized */
|
|
|
|
ext4_ext_mark_uninitialized(&newex);
|
2006-10-11 04:21:03 -04:00
|
|
|
err = ext4_ext_insert_extent(handle, inode, path, &newex);
|
2007-05-24 13:04:25 -04:00
|
|
|
if (err) {
|
|
|
|
/* free data blocks we just allocated */
|
2008-01-29 00:19:52 -05:00
|
|
|
/* not a good idea to call discard here directly,
|
|
|
|
* but otherwise we'd need to call it every free() */
|
|
|
|
ext4_mb_discard_inode_preallocations(inode);
|
2007-05-24 13:04:25 -04:00
|
|
|
ext4_free_blocks(handle, inode, ext_pblock(&newex),
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_ext_get_actual_len(&newex), 0);
|
2006-10-11 04:21:03 -04:00
|
|
|
goto out2;
|
2007-05-24 13:04:25 -04:00
|
|
|
}
|
2006-10-11 04:21:03 -04:00
|
|
|
|
|
|
|
/* previous routine could use block we allocated */
|
2006-10-11 04:21:05 -04:00
|
|
|
newblock = ext_pblock(&newex);
|
2008-01-28 23:58:27 -05:00
|
|
|
allocated = ext4_ext_get_actual_len(&newex);
|
2007-07-17 21:42:38 -04:00
|
|
|
outnew:
|
2008-07-11 19:27:31 -04:00
|
|
|
if (extend_disksize) {
|
|
|
|
disksize = ((loff_t) iblock + ar.len) << inode->i_blkbits;
|
|
|
|
if (disksize > i_size_read(inode))
|
|
|
|
disksize = i_size_read(inode);
|
|
|
|
if (disksize > EXT4_I(inode)->i_disksize)
|
|
|
|
EXT4_I(inode)->i_disksize = disksize;
|
|
|
|
}
|
2008-07-11 19:27:31 -04:00
|
|
|
|
2008-07-11 19:27:31 -04:00
|
|
|
set_buffer_new(bh_result);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
2007-07-17 21:42:41 -04:00
|
|
|
/* Cache only when it is _not_ an uninitialized extent */
|
|
|
|
if (create != EXT4_CREATE_UNINITIALIZED_EXT)
|
|
|
|
ext4_ext_put_in_cache(inode, iblock, allocated, newblock,
|
|
|
|
EXT4_EXT_CACHE_EXTENT);
|
2006-10-11 04:21:03 -04:00
|
|
|
out:
|
|
|
|
if (allocated > max_blocks)
|
|
|
|
allocated = max_blocks;
|
|
|
|
ext4_ext_show_leaf(inode, path);
|
2008-07-11 19:27:31 -04:00
|
|
|
set_buffer_mapped(bh_result);
|
2006-10-11 04:21:03 -04:00
|
|
|
bh_result->b_bdev = inode->i_sb->s_bdev;
|
|
|
|
bh_result->b_blocknr = newblock;
|
|
|
|
out2:
|
|
|
|
if (path) {
|
|
|
|
ext4_ext_drop_refs(path);
|
|
|
|
kfree(path);
|
|
|
|
}
|
|
|
|
return err ? err : allocated;
|
|
|
|
}
|
|
|
|
|
2008-07-11 19:27:31 -04:00
|
|
|
void ext4_ext_truncate(struct inode *inode)
|
2006-10-11 04:21:03 -04:00
|
|
|
{
|
|
|
|
struct address_space *mapping = inode->i_mapping;
|
|
|
|
struct super_block *sb = inode->i_sb;
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t last_block;
|
2006-10-11 04:21:03 -04:00
|
|
|
handle_t *handle;
|
|
|
|
int err = 0;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* probably first extent we're gonna free will be last in block
|
|
|
|
*/
|
|
|
|
err = ext4_writepage_trans_blocks(inode) + 3;
|
|
|
|
handle = ext4_journal_start(inode, err);
|
2008-07-11 19:27:31 -04:00
|
|
|
if (IS_ERR(handle))
|
2006-10-11 04:21:03 -04:00
|
|
|
return;
|
|
|
|
|
2008-07-11 19:27:31 -04:00
|
|
|
if (inode->i_size & (sb->s_blocksize - 1))
|
|
|
|
ext4_block_truncate_page(handle, mapping, inode->i_size);
|
2006-10-11 04:21:03 -04:00
|
|
|
|
2008-07-11 19:27:31 -04:00
|
|
|
if (ext4_orphan_add(handle, inode))
|
|
|
|
goto out_stop;
|
|
|
|
|
2008-01-28 23:58:26 -05:00
|
|
|
down_write(&EXT4_I(inode)->i_data_sem);
|
2006-10-11 04:21:03 -04:00
|
|
|
ext4_ext_invalidate_cache(inode);
|
|
|
|
|
2008-01-29 00:19:52 -05:00
|
|
|
ext4_mb_discard_inode_preallocations(inode);
|
|
|
|
|
2006-10-11 04:21:03 -04:00
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* TODO: optimization is possible here.
|
|
|
|
* Probably we need not scan at all,
|
|
|
|
* because page truncation is enough.
|
2006-10-11 04:21:03 -04:00
|
|
|
*/
|
|
|
|
|
|
|
|
/* we have to know where to truncate from in crash case */
|
|
|
|
EXT4_I(inode)->i_disksize = inode->i_size;
|
|
|
|
ext4_mark_inode_dirty(handle, inode);
|
|
|
|
|
|
|
|
last_block = (inode->i_size + sb->s_blocksize - 1)
|
|
|
|
>> EXT4_BLOCK_SIZE_BITS(sb);
|
|
|
|
err = ext4_ext_remove_space(inode, last_block);
|
|
|
|
|
|
|
|
/* In a multi-transaction truncate, we only make the final
|
2007-07-17 21:42:38 -04:00
|
|
|
* transaction synchronous.
|
|
|
|
*/
|
2006-10-11 04:21:03 -04:00
|
|
|
if (IS_SYNC(inode))
|
|
|
|
handle->h_sync = 1;
|
|
|
|
|
|
|
|
out_stop:
|
2008-07-11 19:27:31 -04:00
|
|
|
up_write(&EXT4_I(inode)->i_data_sem);
|
2006-10-11 04:21:03 -04:00
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* If this was a simple ftruncate() and the file will remain alive,
|
2006-10-11 04:21:03 -04:00
|
|
|
* then we need to clear up the orphan record which we created above.
|
|
|
|
* However, if this was a real unlink then we were called by
|
|
|
|
* ext4_delete_inode(), and we allow that function to clean up the
|
|
|
|
* orphan info for us.
|
|
|
|
*/
|
|
|
|
if (inode->i_nlink)
|
|
|
|
ext4_orphan_del(handle, inode);
|
|
|
|
|
2008-04-29 22:00:41 -04:00
|
|
|
inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
|
|
|
|
ext4_mark_inode_dirty(handle, inode);
|
2006-10-11 04:21:03 -04:00
|
|
|
ext4_journal_stop(handle);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-10-11 04:21:07 -04:00
|
|
|
* ext4_ext_writepage_trans_blocks:
|
|
|
|
* calculate max number of blocks we could modify
|
2006-10-11 04:21:03 -04:00
|
|
|
* in order to allocate new block for an inode
|
|
|
|
*/
|
|
|
|
int ext4_ext_writepage_trans_blocks(struct inode *inode, int num)
|
|
|
|
{
|
|
|
|
int needed;
|
|
|
|
|
|
|
|
needed = ext4_ext_calc_credits_for_insert(inode, NULL);
|
|
|
|
|
2006-10-11 04:21:07 -04:00
|
|
|
/* caller wants to allocate num blocks, but note it includes sb */
|
2006-10-11 04:21:03 -04:00
|
|
|
needed = needed * num - (num - 1);
|
|
|
|
|
|
|
|
#ifdef CONFIG_QUOTA
|
|
|
|
needed += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
return needed;
|
|
|
|
}
|
2007-07-17 21:42:41 -04:00
|
|
|
|
2008-04-29 08:11:12 -04:00
|
|
|
static void ext4_falloc_update_inode(struct inode *inode,
|
|
|
|
int mode, loff_t new_size, int update_ctime)
|
|
|
|
{
|
|
|
|
struct timespec now;
|
|
|
|
|
|
|
|
if (update_ctime) {
|
|
|
|
now = current_fs_time(inode->i_sb);
|
|
|
|
if (!timespec_equal(&inode->i_ctime, &now))
|
|
|
|
inode->i_ctime = now;
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* Update only when preallocation was requested beyond
|
|
|
|
* the file size.
|
|
|
|
*/
|
|
|
|
if (!(mode & FALLOC_FL_KEEP_SIZE) &&
|
|
|
|
new_size > i_size_read(inode)) {
|
|
|
|
i_size_write(inode, new_size);
|
|
|
|
EXT4_I(inode)->i_disksize = new_size;
|
|
|
|
}
|
|
|
|
|
|
|
|
}
|
|
|
|
|
2007-07-17 21:42:41 -04:00
|
|
|
/*
|
|
|
|
* preallocate space for a file. This implements ext4's fallocate inode
|
|
|
|
* operation, which gets called from sys_fallocate system call.
|
|
|
|
* For block-mapped files, posix_fallocate should fall back to the method
|
|
|
|
* of writing zeroes to the required new blocks (the same behavior which is
|
|
|
|
* expected for file systems which do not support fallocate() system call).
|
|
|
|
*/
|
|
|
|
long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len)
|
|
|
|
{
|
|
|
|
handle_t *handle;
|
2008-01-28 23:58:27 -05:00
|
|
|
ext4_lblk_t block;
|
2008-04-29 08:11:12 -04:00
|
|
|
loff_t new_size;
|
2008-01-28 23:58:27 -05:00
|
|
|
unsigned long max_blocks;
|
2007-07-17 21:42:41 -04:00
|
|
|
int ret = 0;
|
|
|
|
int ret2 = 0;
|
|
|
|
int retries = 0;
|
|
|
|
struct buffer_head map_bh;
|
|
|
|
unsigned int credits, blkbits = inode->i_blkbits;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* currently supporting (pre)allocate mode for extent-based
|
|
|
|
* files _only_
|
|
|
|
*/
|
|
|
|
if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL))
|
|
|
|
return -EOPNOTSUPP;
|
|
|
|
|
|
|
|
/* preallocation to directories is currently not supported */
|
|
|
|
if (S_ISDIR(inode->i_mode))
|
|
|
|
return -ENODEV;
|
|
|
|
|
|
|
|
block = offset >> blkbits;
|
2008-04-29 08:11:12 -04:00
|
|
|
/*
|
|
|
|
* We can't just convert len to max_blocks because
|
|
|
|
* If blocksize = 4096 offset = 3072 and len = 2048
|
|
|
|
*/
|
2007-07-17 21:42:41 -04:00
|
|
|
max_blocks = (EXT4_BLOCK_ALIGN(len + offset, blkbits) >> blkbits)
|
2008-04-29 08:11:12 -04:00
|
|
|
- block;
|
2007-07-17 21:42:41 -04:00
|
|
|
/*
|
|
|
|
* credits to insert 1 extent into extent tree + buffers to be able to
|
|
|
|
* modify 1 super block, 1 block bitmap and 1 group descriptor.
|
|
|
|
*/
|
|
|
|
credits = EXT4_DATA_TRANS_BLOCKS(inode->i_sb) + 3;
|
2008-02-15 12:47:21 -05:00
|
|
|
mutex_lock(&inode->i_mutex);
|
2007-07-17 21:42:41 -04:00
|
|
|
retry:
|
|
|
|
while (ret >= 0 && ret < max_blocks) {
|
|
|
|
block = block + ret;
|
|
|
|
max_blocks = max_blocks - ret;
|
|
|
|
handle = ext4_journal_start(inode, credits);
|
|
|
|
if (IS_ERR(handle)) {
|
|
|
|
ret = PTR_ERR(handle);
|
|
|
|
break;
|
|
|
|
}
|
2008-02-15 12:47:21 -05:00
|
|
|
ret = ext4_get_blocks_wrap(handle, inode, block,
|
2007-07-17 21:42:41 -04:00
|
|
|
max_blocks, &map_bh,
|
2008-07-14 17:52:37 -04:00
|
|
|
EXT4_CREATE_UNINITIALIZED_EXT, 0, 0);
|
2008-01-28 23:58:27 -05:00
|
|
|
if (ret <= 0) {
|
2008-02-25 15:41:35 -05:00
|
|
|
#ifdef EXT4FS_DEBUG
|
|
|
|
WARN_ON(ret <= 0);
|
|
|
|
printk(KERN_ERR "%s: ext4_ext_get_blocks "
|
|
|
|
"returned error inode#%lu, block=%u, "
|
|
|
|
"max_blocks=%lu", __func__,
|
2008-01-28 23:58:27 -05:00
|
|
|
inode->i_ino, block, max_blocks);
|
2008-02-25 15:41:35 -05:00
|
|
|
#endif
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_mark_inode_dirty(handle, inode);
|
|
|
|
ret2 = ext4_journal_stop(handle);
|
|
|
|
break;
|
|
|
|
}
|
2008-04-29 08:11:12 -04:00
|
|
|
if ((block + ret) >= (EXT4_BLOCK_ALIGN(offset + len,
|
|
|
|
blkbits) >> blkbits))
|
|
|
|
new_size = offset + len;
|
|
|
|
else
|
|
|
|
new_size = (block + ret) << blkbits;
|
2007-07-17 21:42:41 -04:00
|
|
|
|
2008-04-29 08:11:12 -04:00
|
|
|
ext4_falloc_update_inode(inode, mode, new_size,
|
|
|
|
buffer_new(&map_bh));
|
2007-07-17 21:42:41 -04:00
|
|
|
ext4_mark_inode_dirty(handle, inode);
|
|
|
|
ret2 = ext4_journal_stop(handle);
|
|
|
|
if (ret2)
|
|
|
|
break;
|
|
|
|
}
|
2008-04-29 08:11:12 -04:00
|
|
|
if (ret == -ENOSPC &&
|
|
|
|
ext4_should_retry_alloc(inode->i_sb, &retries)) {
|
|
|
|
ret = 0;
|
2007-07-17 21:42:41 -04:00
|
|
|
goto retry;
|
|
|
|
}
|
2008-02-15 12:47:21 -05:00
|
|
|
mutex_unlock(&inode->i_mutex);
|
2007-07-17 21:42:41 -04:00
|
|
|
return ret > 0 ? ret2 : ret;
|
|
|
|
}
|