aboutsummaryrefslogtreecommitdiffstats
path: root/fs/ext4/extents.c
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ext4/extents.c')
-rw-r--r--fs/ext4/extents.c198
1 files changed, 97 insertions, 101 deletions
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 0ecf819bf189..f1956288307f 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -27,8 +27,8 @@
#include <linux/slab.h>
#include <linux/uaccess.h>
#include <linux/fiemap.h>
-#include <linux/backing-dev.h>
#include <linux/iomap.h>
+#include <linux/sched/mm.h>
#include "ext4_jbd2.h"
#include "ext4_extents.h"
#include "xattr.h"
@@ -97,7 +97,7 @@ static int ext4_ext_trunc_restart_fn(struct inode *inode, int *dropped)
* Drop i_data_sem to avoid deadlock with ext4_map_blocks. At this
* moment, get_block can be called only for blocks inside i_size since
* page cache has been already dropped and writes are blocked by
- * i_mutex. So we can safely drop the i_data_sem here.
+ * i_rwsem. So we can safely drop the i_data_sem here.
*/
BUG_ON(EXT4_JOURNAL(inode) == NULL);
ext4_discard_preallocations(inode, 0);
@@ -106,6 +106,25 @@ static int ext4_ext_trunc_restart_fn(struct inode *inode, int *dropped)
return 0;
}
+static void ext4_ext_drop_refs(struct ext4_ext_path *path)
+{
+ int depth, i;
+
+ if (!path)
+ return;
+ depth = path->p_depth;
+ for (i = 0; i <= depth; i++, path++) {
+ brelse(path->p_bh);
+ path->p_bh = NULL;
+ }
+}
+
+void ext4_free_ext_path(struct ext4_ext_path *path)
+{
+ ext4_ext_drop_refs(path);
+ kfree(path);
+}
+
/*
* Make sure 'handle' has at least 'check_cred' credits. If not, restart
* transaction with 'restart_cred' credits. The function drops i_data_sem
@@ -372,7 +391,7 @@ static int ext4_valid_extent_entries(struct inode *inode,
{
unsigned short entries;
ext4_lblk_t lblock = 0;
- ext4_lblk_t prev = 0;
+ ext4_lblk_t cur = 0;
if (eh->eh_entries == 0)
return 1;
@@ -396,11 +415,11 @@ static int ext4_valid_extent_entries(struct inode *inode,
/* Check for overlapping extents */
lblock = le32_to_cpu(ext->ee_block);
- if ((lblock <= prev) && prev) {
+ if (lblock < cur) {
*pblk = ext4_ext_pblock(ext);
return 0;
}
- prev = lblock + ext4_ext_get_actual_len(ext) - 1;
+ cur = lblock + ext4_ext_get_actual_len(ext);
ext++;
entries--;
}
@@ -420,13 +439,13 @@ static int ext4_valid_extent_entries(struct inode *inode,
/* Check for overlapping index extents */
lblock = le32_to_cpu(ext_idx->ei_block);
- if ((lblock <= prev) && prev) {
+ if (lblock < cur) {
*pblk = ext4_idx_pblock(ext_idx);
return 0;
}
ext_idx++;
entries--;
- prev = lblock;
+ cur = lblock + 1;
}
}
return 1;
@@ -460,6 +479,10 @@ static int __ext4_ext_check(const char *function, unsigned int line,
error_msg = "invalid eh_entries";
goto corrupted;
}
+ if (unlikely((eh->eh_entries == 0) && (depth > 0))) {
+ error_msg = "eh_entries is 0 but eh_depth is > 0";
+ goto corrupted;
+ }
if (!ext4_valid_extent_entries(inode, eh, lblk, &pblk, depth)) {
error_msg = "invalid extent entries";
goto corrupted;
@@ -632,8 +655,7 @@ int ext4_ext_precache(struct inode *inode)
ext4_set_inode_state(inode, EXT4_STATE_EXT_PRECACHED);
out:
up_read(&ei->i_data_sem);
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
return ret;
}
@@ -720,19 +742,6 @@ static void ext4_ext_show_move(struct inode *inode, struct ext4_ext_path *path,
#define ext4_ext_show_move(inode, path, newblock, level)
#endif
-void ext4_ext_drop_refs(struct ext4_ext_path *path)
-{
- int depth, i;
-
- if (!path)
- return;
- depth = path->p_depth;
- for (i = 0; i <= depth; i++, path++) {
- brelse(path->p_bh);
- path->p_bh = NULL;
- }
-}
-
/*
* ext4_ext_binsearch_idx:
* binary search for the closest index of the given block
@@ -951,8 +960,7 @@ ext4_find_extent(struct inode *inode, ext4_lblk_t block,
return path;
err:
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
if (orig_path)
*orig_path = NULL;
return ERR_PTR(ret);
@@ -1496,8 +1504,7 @@ static int ext4_ext_search_left(struct inode *inode,
EXT4_ERROR_INODE(inode,
"ix (%d) != EXT_FIRST_INDEX (%d) (depth %d)!",
ix != NULL ? le32_to_cpu(ix->ei_block) : 0,
- EXT_FIRST_INDEX(path[depth].p_hdr) != NULL ?
- le32_to_cpu(EXT_FIRST_INDEX(path[depth].p_hdr)->ei_block) : 0,
+ le32_to_cpu(EXT_FIRST_INDEX(path[depth].p_hdr)->ei_block),
depth);
return -EFSCORRUPTED;
}
@@ -2025,7 +2032,6 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode,
+ ext4_ext_get_actual_len(newext));
if (unwritten)
ext4_ext_mark_unwritten(ex);
- eh = path[depth].p_hdr;
nearex = ex;
goto merge;
}
@@ -2054,7 +2060,6 @@ prepend:
+ ext4_ext_get_actual_len(newext));
if (unwritten)
ext4_ext_mark_unwritten(ex);
- eh = path[depth].p_hdr;
nearex = ex;
goto merge;
}
@@ -2173,8 +2178,7 @@ merge:
err = ext4_ext_dirty(handle, inode, path + path->p_depth);
cleanup:
- ext4_ext_drop_refs(npath);
- kfree(npath);
+ ext4_free_ext_path(npath);
return err;
}
@@ -3060,8 +3064,7 @@ again:
}
}
out:
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
path = NULL;
if (err == -EAGAIN)
goto again;
@@ -3371,7 +3374,6 @@ static int ext4_split_extent(handle_t *handle,
return -EFSCORRUPTED;
}
unwritten = ext4_ext_is_unwritten(ex);
- split_flag1 = 0;
if (map->m_lblk >= ee_block) {
split_flag1 = split_flag & EXT4_EXT_DATA_VALID2;
@@ -4375,8 +4377,7 @@ got_allocated_blocks:
allocated = map->m_len;
ext4_ext_show_leaf(inode, path);
out:
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
trace_ext4_ext_map_blocks_exit(inode, flags, map,
err ? err : allocated);
@@ -4407,8 +4408,7 @@ retry:
err = ext4_es_remove_extent(inode, last_block,
EXT_MAX_BLOCKS - last_block);
if (err == -ENOMEM) {
- cond_resched();
- congestion_wait(BLK_RW_ASYNC, HZ/50);
+ memalloc_retry_wait(GFP_ATOMIC);
goto retry;
}
if (err)
@@ -4416,8 +4416,7 @@ retry:
retry_remove_space:
err = ext4_ext_remove_space(inode, last_block, EXT_MAX_BLOCKS - 1);
if (err == -ENOMEM) {
- cond_resched();
- congestion_wait(BLK_RW_ASYNC, HZ/50);
+ memalloc_retry_wait(GFP_ATOMIC);
goto retry_remove_space;
}
return err;
@@ -4506,9 +4505,9 @@ retry:
return ret > 0 ? ret2 : ret;
}
-static int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len);
+static int ext4_collapse_range(struct file *file, loff_t offset, loff_t len);
-static int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len);
+static int ext4_insert_range(struct file *file, loff_t offset, loff_t len);
static long ext4_zero_range(struct file *file, loff_t offset,
loff_t len, int mode)
@@ -4577,9 +4576,13 @@ static long ext4_zero_range(struct file *file, loff_t offset,
flags = EXT4_GET_BLOCKS_CREATE_UNWRIT_EXT;
- /* Wait all existing dio workers, newcomers will block on i_mutex */
+ /* Wait all existing dio workers, newcomers will block on i_rwsem */
inode_dio_wait(inode);
+ ret = file_modified(file);
+ if (ret)
+ goto out_mutex;
+
/* Preallocate the range including the unaligned edges */
if (partial_begin || partial_end) {
ret = ext4_alloc_file_blocks(file,
@@ -4647,8 +4650,6 @@ static long ext4_zero_range(struct file *file, loff_t offset,
ret = ext4_mark_inode_dirty(handle, inode);
if (unlikely(ret))
goto out_handle;
- ext4_fc_track_range(handle, inode, offset >> inode->i_sb->s_blocksize_bits,
- (offset + len - 1) >> inode->i_sb->s_blocksize_bits);
/* Zero out partial block at the edges of the range */
ret = ext4_zero_partial_blocks(handle, inode, offset, len);
if (ret >= 0)
@@ -4697,24 +4698,24 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
FALLOC_FL_INSERT_RANGE))
return -EOPNOTSUPP;
- ext4_fc_start_update(inode);
+ inode_lock(inode);
+ ret = ext4_convert_inline_data(inode);
+ inode_unlock(inode);
+ if (ret)
+ goto exit;
if (mode & FALLOC_FL_PUNCH_HOLE) {
- ret = ext4_punch_hole(inode, offset, len);
+ ret = ext4_punch_hole(file, offset, len);
goto exit;
}
- ret = ext4_convert_inline_data(inode);
- if (ret)
- goto exit;
-
if (mode & FALLOC_FL_COLLAPSE_RANGE) {
- ret = ext4_collapse_range(inode, offset, len);
+ ret = ext4_collapse_range(file, offset, len);
goto exit;
}
if (mode & FALLOC_FL_INSERT_RANGE) {
- ret = ext4_insert_range(inode, offset, len);
+ ret = ext4_insert_range(file, offset, len);
goto exit;
}
@@ -4747,9 +4748,13 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
goto out;
}
- /* Wait all existing dio workers, newcomers will block on i_mutex */
+ /* Wait all existing dio workers, newcomers will block on i_rwsem */
inode_dio_wait(inode);
+ ret = file_modified(file);
+ if (ret)
+ goto out;
+
ret = ext4_alloc_file_blocks(file, lblk, max_blocks, new_size, flags);
if (ret)
goto out;
@@ -4762,7 +4767,6 @@ out:
inode_unlock(inode);
trace_ext4_fallocate_exit(inode, offset, max_blocks, ret);
exit:
- ext4_fc_stop_update(inode);
return ret;
}
@@ -5242,8 +5246,7 @@ again:
break;
}
out:
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
return ret;
}
@@ -5252,8 +5255,9 @@ out:
* This implements the fallocate's collapse range functionality for ext4
* Returns: 0 and non-zero on error.
*/
-static int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
+static int ext4_collapse_range(struct file *file, loff_t offset, loff_t len)
{
+ struct inode *inode = file_inode(file);
struct super_block *sb = inode->i_sb;
struct address_space *mapping = inode->i_mapping;
ext4_lblk_t punch_start, punch_stop;
@@ -5305,6 +5309,10 @@ static int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
/* Wait for existing dio to complete */
inode_dio_wait(inode);
+ ret = file_modified(file);
+ if (ret)
+ goto out_mutex;
+
/*
* Prevent page faults from reinstantiating pages we have released from
* page cache.
@@ -5344,7 +5352,7 @@ static int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
ret = PTR_ERR(handle);
goto out_mmap;
}
- ext4_fc_start_ineligible(sb, EXT4_FC_REASON_FALLOC_RANGE);
+ ext4_fc_mark_ineligible(sb, EXT4_FC_REASON_FALLOC_RANGE, handle);
down_write(&EXT4_I(inode)->i_data_sem);
ext4_discard_preallocations(inode, 0);
@@ -5383,7 +5391,6 @@ static int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
out_stop:
ext4_journal_stop(handle);
- ext4_fc_stop_ineligible(sb);
out_mmap:
filemap_invalidate_unlock(mapping);
out_mutex:
@@ -5399,8 +5406,9 @@ out_mutex:
* by len bytes.
* Returns 0 on success, error otherwise.
*/
-static int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
+static int ext4_insert_range(struct file *file, loff_t offset, loff_t len)
{
+ struct inode *inode = file_inode(file);
struct super_block *sb = inode->i_sb;
struct address_space *mapping = inode->i_mapping;
handle_t *handle;
@@ -5457,6 +5465,10 @@ static int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
/* Wait for existing dio to complete */
inode_dio_wait(inode);
+ ret = file_modified(file);
+ if (ret)
+ goto out_mutex;
+
/*
* Prevent page faults from reinstantiating pages we have released from
* page cache.
@@ -5485,7 +5497,7 @@ static int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
ret = PTR_ERR(handle);
goto out_mmap;
}
- ext4_fc_start_ineligible(sb, EXT4_FC_REASON_FALLOC_RANGE);
+ ext4_fc_mark_ineligible(sb, EXT4_FC_REASON_FALLOC_RANGE, handle);
/* Expand file to avoid data loss if there is error while shifting */
inode->i_size += len;
@@ -5526,15 +5538,13 @@ static int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
EXT4_GET_BLOCKS_METADATA_NOFAIL);
}
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
if (ret < 0) {
up_write(&EXT4_I(inode)->i_data_sem);
goto out_stop;
}
} else {
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
}
ret = ext4_es_remove_extent(inode, offset_lblk,
@@ -5560,7 +5570,6 @@ static int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
out_stop:
ext4_journal_stop(handle);
- ext4_fc_stop_ineligible(sb);
out_mmap:
filemap_invalidate_unlock(mapping);
out_mutex:
@@ -5583,7 +5592,7 @@ out_mutex:
* stuff such as page-cache locking consistency, bh mapping consistency or
* extent's data copying must be performed by caller.
* Locking:
- * i_mutex is held for both inodes
+ * i_rwsem is held for both inodes
* i_data_sem is locked for write for both inodes
* Assumptions:
* All pages from requested range are locked for both inodes
@@ -5755,10 +5764,8 @@ ext4_swap_extents(handle_t *handle, struct inode *inode1,
count -= len;
repeat:
- ext4_ext_drop_refs(path1);
- kfree(path1);
- ext4_ext_drop_refs(path2);
- kfree(path2);
+ ext4_free_ext_path(path1);
+ ext4_free_ext_path(path2);
path1 = path2 = NULL;
}
return replaced_count;
@@ -5837,8 +5844,7 @@ int ext4_clu_mapped(struct inode *inode, ext4_lblk_t lclu)
}
out:
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
return err ? err : mapped;
}
@@ -5905,8 +5911,7 @@ int ext4_ext_replay_update_ex(struct inode *inode, ext4_lblk_t start,
ret = ext4_ext_dirty(NULL, inode, &path[path->p_depth]);
up_write(&EXT4_I(inode)->i_data_sem);
out:
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
ext4_mark_inode_dirty(NULL, inode);
return ret;
}
@@ -5924,8 +5929,7 @@ void ext4_ext_replay_shrink_inode(struct inode *inode, ext4_lblk_t end)
return;
ex = path[path->p_depth].p_ext;
if (!ex) {
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
ext4_mark_inode_dirty(NULL, inode);
return;
}
@@ -5938,8 +5942,7 @@ void ext4_ext_replay_shrink_inode(struct inode *inode, ext4_lblk_t end)
ext4_ext_dirty(NULL, inode, &path[path->p_depth]);
up_write(&EXT4_I(inode)->i_data_sem);
ext4_mark_inode_dirty(NULL, inode);
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
}
}
@@ -5978,13 +5981,11 @@ int ext4_ext_replay_set_iblocks(struct inode *inode)
return PTR_ERR(path);
ex = path[path->p_depth].p_ext;
if (!ex) {
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
goto out;
}
end = le32_to_cpu(ex->ee_block) + ext4_ext_get_actual_len(ex);
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
/* Count the number of data blocks */
cur = 0;
@@ -6014,30 +6015,26 @@ int ext4_ext_replay_set_iblocks(struct inode *inode)
if (IS_ERR(path))
goto out;
numblks += path->p_depth;
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
while (cur < end) {
path = ext4_find_extent(inode, cur, NULL, 0);
if (IS_ERR(path))
break;
ex = path[path->p_depth].p_ext;
if (!ex) {
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
return 0;
}
cur = max(cur + 1, le32_to_cpu(ex->ee_block) +
ext4_ext_get_actual_len(ex));
ret = skip_hole(inode, &cur);
if (ret < 0) {
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
break;
}
path2 = ext4_find_extent(inode, cur, NULL, 0);
if (IS_ERR(path2)) {
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
break;
}
for (i = 0; i <= max(path->p_depth, path2->p_depth); i++) {
@@ -6051,10 +6048,8 @@ int ext4_ext_replay_set_iblocks(struct inode *inode)
if (cmp1 != cmp2 && cmp2 != 0)
numblks++;
}
- ext4_ext_drop_refs(path);
- ext4_ext_drop_refs(path2);
- kfree(path);
- kfree(path2);
+ ext4_free_ext_path(path);
+ ext4_free_ext_path(path2);
}
out:
@@ -6081,13 +6076,11 @@ int ext4_ext_clear_bb(struct inode *inode)
return PTR_ERR(path);
ex = path[path->p_depth].p_ext;
if (!ex) {
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
return 0;
}
end = le32_to_cpu(ex->ee_block) + ext4_ext_get_actual_len(ex);
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
cur = 0;
while (cur < end) {
@@ -6103,11 +6096,14 @@ int ext4_ext_clear_bb(struct inode *inode)
ext4_mb_mark_bb(inode->i_sb,
path[j].p_block, 1, 0);
+ ext4_fc_record_regions(inode->i_sb, inode->i_ino,
+ 0, path[j].p_block, 1, 1);
}
- ext4_ext_drop_refs(path);
- kfree(path);
+ ext4_free_ext_path(path);
}
ext4_mb_mark_bb(inode->i_sb, map.m_pblk, map.m_len, 0);
+ ext4_fc_record_regions(inode->i_sb, inode->i_ino,
+ map.m_lblk, map.m_pblk, map.m_len, 1);
}
cur = cur + map.m_len;
}
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.