openkylin
/
linux
mirror of https://gitee.com/openkylin/linux.git
9
0
Fork 0
Code Issues Projects Releases Wiki Activity
linux/fs/f2fs/inode.c

291 lines
7.6 KiB
C
Raw Normal View History

f2fs: adjust kernel coding style As pointed out by Randy Dunlap, this patch removes all usage of "/**" for comment blocks. Instead, just use "/*". Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-29 12:28:09 +08:00
/*
f2fs: add core inode operations This adds core functions to get, read, write, and evict an inode. Signed-off-by: Changman Lee <cm224.lee@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-02 16:10:40 +08:00
* fs/f2fs/inode.c
*
* Copyright (c) 2012 Samsung Electronics Co., Ltd.
* http://www.samsung.com/
*
* 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.
*/
#include <linux/fs.h>
#include <linux/f2fs_fs.h>
#include <linux/buffer_head.h>
#include <linux/writeback.h>
#include "f2fs.h"
#include "node.h"
struct f2fs_iget_args {
u64 ino;
int on_free;
};
void f2fs_set_inode_flags(struct inode *inode)
{
unsigned int flags = F2FS_I(inode)->i_flags;
inode->i_flags &= ~(S_SYNC | S_APPEND | S_IMMUTABLE |
S_NOATIME | S_DIRSYNC);
if (flags & FS_SYNC_FL)
inode->i_flags |= S_SYNC;
if (flags & FS_APPEND_FL)
inode->i_flags |= S_APPEND;
if (flags & FS_IMMUTABLE_FL)
inode->i_flags |= S_IMMUTABLE;
if (flags & FS_NOATIME_FL)
inode->i_flags |= S_NOATIME;
if (flags & FS_DIRSYNC_FL)
inode->i_flags |= S_DIRSYNC;
}
static int f2fs_iget_test(struct inode *inode, void *data)
{
struct f2fs_iget_args *args = data;
if (inode->i_ino != args->ino)
return 0;
if (inode->i_state & (I_FREEING | I_WILL_FREE)) {
args->on_free = 1;
return 0;
}
return 1;
}
struct inode *f2fs_iget_nowait(struct super_block *sb, unsigned long ino)
{
struct f2fs_iget_args args = {
.ino = ino,
.on_free = 0
};
struct inode *inode = ilookup5(sb, ino, f2fs_iget_test, &args);
if (inode)
return inode;
if (!args.on_free)
return f2fs_iget(sb, ino);
return ERR_PTR(-ENOENT);
}
static int do_read_inode(struct inode *inode)
{
struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
struct f2fs_inode_info *fi = F2FS_I(inode);
struct page *node_page;
struct f2fs_node *rn;
struct f2fs_inode *ri;
/* Check if ino is within scope */
check_nid_range(sbi, inode->i_ino);
node_page = get_node_page(sbi, inode->i_ino);
if (IS_ERR(node_page))
return PTR_ERR(node_page);
rn = page_address(node_page);
ri = &(rn->i);
inode->i_mode = le16_to_cpu(ri->i_mode);
i_uid_write(inode, le32_to_cpu(ri->i_uid));
i_gid_write(inode, le32_to_cpu(ri->i_gid));
set_nlink(inode, le32_to_cpu(ri->i_links));
inode->i_size = le64_to_cpu(ri->i_size);
inode->i_blocks = le64_to_cpu(ri->i_blocks);
inode->i_atime.tv_sec = le64_to_cpu(ri->i_atime);
inode->i_ctime.tv_sec = le64_to_cpu(ri->i_ctime);
inode->i_mtime.tv_sec = le64_to_cpu(ri->i_mtime);
inode->i_atime.tv_nsec = le32_to_cpu(ri->i_atime_nsec);
inode->i_ctime.tv_nsec = le32_to_cpu(ri->i_ctime_nsec);
inode->i_mtime.tv_nsec = le32_to_cpu(ri->i_mtime_nsec);
inode->i_generation = le32_to_cpu(ri->i_generation);
f2fs: save device node number into f2fs_inode This patch stores inode->i_rdev into on-disk inode structure. Alun reported that: aspire tmp # mount -t f2fs /dev/sdb mnt aspire tmp # mknod mnt/sda1 b 8 1 aspire tmp # mknod mnt/null c 1 3 aspire tmp # mknod mnt/console c 5 1 aspire tmp # ls -l mnt total 2 crw-r--r-- 1 root root 5, 1 Jan 22 18:44 console crw-r--r-- 1 root root 1, 3 Jan 22 18:44 null brw-r--r-- 1 root root 8, 1 Jan 22 18:44 sda1 aspire tmp # umount mnt aspire tmp # mount -t f2fs /dev/sdb mnt aspire tmp # ls -l mnt total 2 crw-r--r-- 1 root root 0, 0 Jan 22 18:44 console crw-r--r-- 1 root root 0, 0 Jan 22 18:44 null brw-r--r-- 1 root root 0, 0 Jan 22 18:44 sda1 In this report, f2fs lost the major/minor numbers of device files after umount. The reason was revealed that f2fs does not store the inode->i_rdev to the on-disk inode data structure. So, as the other file systems do, f2fs also stores i_rdev into the i_addr fields in on-disk inode structure without any on-disk layout changes. Note that, this bug is limited to device files made by mknod(). Reported-and-Tested-by: Alun Jones <alun.linux@ty-penguin.org.uk> Signed-off-by: Changman Lee <cm224.lee@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-01-23 08:40:23 +08:00
if (ri->i_addr[0])
inode->i_rdev = old_decode_dev(le32_to_cpu(ri->i_addr[0]));
else
inode->i_rdev = new_decode_dev(le32_to_cpu(ri->i_addr[1]));
f2fs: add core inode operations This adds core functions to get, read, write, and evict an inode. Signed-off-by: Changman Lee <cm224.lee@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-02 16:10:40 +08:00
fi->i_current_depth = le32_to_cpu(ri->i_current_depth);
fi->i_xattr_nid = le32_to_cpu(ri->i_xattr_nid);
fi->i_flags = le32_to_cpu(ri->i_flags);
fi->flags = 0;
fi->data_version = le64_to_cpu(F2FS_CKPT(sbi)->checkpoint_ver) - 1;
fi->i_advise = ri->i_advise;
f2fs: fix tracking parent inode number Previously, f2fs didn't track the parent inode number correctly which is stored in each f2fs_inode. In the case of the following scenario, a bug can be occured. Let's suppose there are one directory, "/b", and two files, "/a" and "/b/a". - pino of "/a" is ROOT_INO. - pino of "/b/a" is DIR_B_INO. Then, # sync : The inode pages of "/a" and "/b/a" contain the parent inode numbers as ROOT_INO and DIR_B_INO respectively. # mv /a /b/a : The parent inode number of "/a" should be changed to DIR_B_INO, but f2fs didn't do that. Ref. f2fs_set_link(). In order to fix this clearly, I added i_pino in f2fs_inode_info, and whenever it needs to be changed like in f2fs_add_link() and f2fs_set_link(), it is updated temporarily in f2fs_inode_info. And later, f2fs_write_inode() stores the latest information to the inode pages. For power-off-recovery, f2fs_sync_file() triggers simply f2fs_write_inode(). Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-12-10 16:52:48 +08:00
fi->i_pino = le32_to_cpu(ri->i_pino);
f2fs: add core inode operations This adds core functions to get, read, write, and evict an inode. Signed-off-by: Changman Lee <cm224.lee@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-02 16:10:40 +08:00
get_extent_info(&fi->ext, ri->i_ext);
f2fs_put_page(node_page, 1);
return 0;
}
struct inode *f2fs_iget(struct super_block *sb, unsigned long ino)
{
struct f2fs_sb_info *sbi = F2FS_SB(sb);
struct inode *inode;
int ret;
inode = iget_locked(sb, ino);
if (!inode)
return ERR_PTR(-ENOMEM);
if (!(inode->i_state & I_NEW))
return inode;
if (ino == F2FS_NODE_INO(sbi) || ino == F2FS_META_INO(sbi))
goto make_now;
ret = do_read_inode(inode);
if (ret)
goto bad_inode;
if (!sbi->por_doing && inode->i_nlink == 0) {
ret = -ENOENT;
goto bad_inode;
}
make_now:
if (ino == F2FS_NODE_INO(sbi)) {
inode->i_mapping->a_ops = &f2fs_node_aops;
mapping_set_gfp_mask(inode->i_mapping, GFP_F2FS_ZERO);
} else if (ino == F2FS_META_INO(sbi)) {
inode->i_mapping->a_ops = &f2fs_meta_aops;
mapping_set_gfp_mask(inode->i_mapping, GFP_F2FS_ZERO);
} else if (S_ISREG(inode->i_mode)) {
inode->i_op = &f2fs_file_inode_operations;
inode->i_fop = &f2fs_file_operations;
inode->i_mapping->a_ops = &f2fs_dblock_aops;
} else if (S_ISDIR(inode->i_mode)) {
inode->i_op = &f2fs_dir_inode_operations;
inode->i_fop = &f2fs_dir_operations;
inode->i_mapping->a_ops = &f2fs_dblock_aops;
mapping_set_gfp_mask(inode->i_mapping, GFP_HIGHUSER_MOVABLE |
__GFP_ZERO);
} else if (S_ISLNK(inode->i_mode)) {
inode->i_op = &f2fs_symlink_inode_operations;
inode->i_mapping->a_ops = &f2fs_dblock_aops;
} else if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode) ||
S_ISFIFO(inode->i_mode) || S_ISSOCK(inode->i_mode)) {
inode->i_op = &f2fs_special_inode_operations;
init_special_inode(inode, inode->i_mode, inode->i_rdev);
} else {
ret = -EIO;
goto bad_inode;
}
unlock_new_inode(inode);
return inode;
bad_inode:
iget_failed(inode);
return ERR_PTR(ret);
}
void update_inode(struct inode *inode, struct page *node_page)
{
struct f2fs_node *rn;
struct f2fs_inode *ri;
wait_on_page_writeback(node_page);
rn = page_address(node_page);
ri = &(rn->i);
ri->i_mode = cpu_to_le16(inode->i_mode);
ri->i_advise = F2FS_I(inode)->i_advise;
ri->i_uid = cpu_to_le32(i_uid_read(inode));
ri->i_gid = cpu_to_le32(i_gid_read(inode));
ri->i_links = cpu_to_le32(inode->i_nlink);
ri->i_size = cpu_to_le64(i_size_read(inode));
ri->i_blocks = cpu_to_le64(inode->i_blocks);
set_raw_extent(&F2FS_I(inode)->ext, &ri->i_ext);
ri->i_atime = cpu_to_le64(inode->i_atime.tv_sec);
ri->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
ri->i_mtime = cpu_to_le64(inode->i_mtime.tv_sec);
ri->i_atime_nsec = cpu_to_le32(inode->i_atime.tv_nsec);
ri->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
ri->i_mtime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
ri->i_current_depth = cpu_to_le32(F2FS_I(inode)->i_current_depth);
ri->i_xattr_nid = cpu_to_le32(F2FS_I(inode)->i_xattr_nid);
ri->i_flags = cpu_to_le32(F2FS_I(inode)->i_flags);
f2fs: fix tracking parent inode number Previously, f2fs didn't track the parent inode number correctly which is stored in each f2fs_inode. In the case of the following scenario, a bug can be occured. Let's suppose there are one directory, "/b", and two files, "/a" and "/b/a". - pino of "/a" is ROOT_INO. - pino of "/b/a" is DIR_B_INO. Then, # sync : The inode pages of "/a" and "/b/a" contain the parent inode numbers as ROOT_INO and DIR_B_INO respectively. # mv /a /b/a : The parent inode number of "/a" should be changed to DIR_B_INO, but f2fs didn't do that. Ref. f2fs_set_link(). In order to fix this clearly, I added i_pino in f2fs_inode_info, and whenever it needs to be changed like in f2fs_add_link() and f2fs_set_link(), it is updated temporarily in f2fs_inode_info. And later, f2fs_write_inode() stores the latest information to the inode pages. For power-off-recovery, f2fs_sync_file() triggers simply f2fs_write_inode(). Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-12-10 16:52:48 +08:00
ri->i_pino = cpu_to_le32(F2FS_I(inode)->i_pino);
f2fs: add core inode operations This adds core functions to get, read, write, and evict an inode. Signed-off-by: Changman Lee <cm224.lee@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-02 16:10:40 +08:00
ri->i_generation = cpu_to_le32(inode->i_generation);
f2fs: save device node number into f2fs_inode This patch stores inode->i_rdev into on-disk inode structure. Alun reported that: aspire tmp # mount -t f2fs /dev/sdb mnt aspire tmp # mknod mnt/sda1 b 8 1 aspire tmp # mknod mnt/null c 1 3 aspire tmp # mknod mnt/console c 5 1 aspire tmp # ls -l mnt total 2 crw-r--r-- 1 root root 5, 1 Jan 22 18:44 console crw-r--r-- 1 root root 1, 3 Jan 22 18:44 null brw-r--r-- 1 root root 8, 1 Jan 22 18:44 sda1 aspire tmp # umount mnt aspire tmp # mount -t f2fs /dev/sdb mnt aspire tmp # ls -l mnt total 2 crw-r--r-- 1 root root 0, 0 Jan 22 18:44 console crw-r--r-- 1 root root 0, 0 Jan 22 18:44 null brw-r--r-- 1 root root 0, 0 Jan 22 18:44 sda1 In this report, f2fs lost the major/minor numbers of device files after umount. The reason was revealed that f2fs does not store the inode->i_rdev to the on-disk inode data structure. So, as the other file systems do, f2fs also stores i_rdev into the i_addr fields in on-disk inode structure without any on-disk layout changes. Note that, this bug is limited to device files made by mknod(). Reported-and-Tested-by: Alun Jones <alun.linux@ty-penguin.org.uk> Signed-off-by: Changman Lee <cm224.lee@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-01-23 08:40:23 +08:00
if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) {
if (old_valid_dev(inode->i_rdev)) {
ri->i_addr[0] =
cpu_to_le32(old_encode_dev(inode->i_rdev));
ri->i_addr[1] = 0;
} else {
ri->i_addr[0] = 0;
ri->i_addr[1] =
cpu_to_le32(new_encode_dev(inode->i_rdev));
ri->i_addr[2] = 0;
}
}
f2fs: fix handling errors got by f2fs_write_inode Ruslan reported that f2fs hangs with an infinite loop in f2fs_sync_file(): while (sync_node_pages(sbi, inode->i_ino, &wbc) == 0) f2fs_write_inode(inode, NULL); The reason was revealed that the cold flag is not set even thought this inode is a normal file. Therefore, sync_node_pages() skips to write node blocks since it only writes cold node blocks. The cold flag is stored to the node_footer in node block, and whenever a new node page is allocated, it is set according to its file type, file or directory. But, after sudden-power-off, when recovering the inode page, f2fs doesn't recover its cold flag. So, let's assign the cold flag in more right places. One more thing: If f2fs_write_inode() returns an error due to whatever situations, there would be no dirty node pages so that sync_node_pages() returns zero. (i.e., zero means nothing was written.) Reported-by: Ruslan N. Marchenko <me@ruff.mobi> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-12-19 14:28:39 +08:00
set_cold_node(inode, node_page);
f2fs: add core inode operations This adds core functions to get, read, write, and evict an inode. Signed-off-by: Changman Lee <cm224.lee@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-02 16:10:40 +08:00
set_page_dirty(node_page);
}
int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)
{
struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
struct page *node_page;
bool need_lock = false;
if (inode->i_ino == F2FS_NODE_INO(sbi) ||
inode->i_ino == F2FS_META_INO(sbi))
return 0;
f2fs: add f2fs_balance_fs in several interfaces The f2fs_balance_fs() is to check the number of free sections and decide whether it needs to conduct cleaning or not. If there are not enough free sections, the cleaning job should be started. In order to control an amount of free sections even under high utilization, f2fs should call f2fs_balance_fs at all the VFS interfaces that are able to produce dirty pages. This patch adds the function calls in the missing interfaces as follows. 1. f2fs_setxattr() The f2fs_setxattr() produces dirty node pages so that we should call f2fs_balance_fs() either likewise doing in other VFS interfaces such as f2fs_lookup(), f2fs_mkdir(), and so on. 2. f2fs_sync_file() We should guarantee serving free sections for syncing metadata during fsync. Previously, there is no space check before triggering checkpoint and sync_node_pages. Therefore, if a bunch of fsync calls are triggered under 100% of FS utilization, f2fs is able to be faced with no free sections, resulting in BUG_ON(). 3. f2fs_sync_fs() Before calling write_checkpoint(), we should guarantee that there are minimum free sections. 4. f2fs_write_inode() f2fs_write_inode() is also able to produce dirty node pages. Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-01-11 12:10:49 +08:00
if (wbc)
f2fs_balance_fs(sbi);
f2fs: add core inode operations This adds core functions to get, read, write, and evict an inode. Signed-off-by: Changman Lee <cm224.lee@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-02 16:10:40 +08:00
node_page = get_node_page(sbi, inode->i_ino);
if (IS_ERR(node_page))
return PTR_ERR(node_page);
if (!PageDirty(node_page)) {
need_lock = true;
f2fs_put_page(node_page, 1);
mutex_lock(&sbi->write_inode);
node_page = get_node_page(sbi, inode->i_ino);
if (IS_ERR(node_page)) {
mutex_unlock(&sbi->write_inode);
return PTR_ERR(node_page);
}
}
update_inode(inode, node_page);
f2fs_put_page(node_page, 1);
if (need_lock)
mutex_unlock(&sbi->write_inode);
return 0;
}
f2fs: adjust kernel coding style As pointed out by Randy Dunlap, this patch removes all usage of "/**" for comment blocks. Instead, just use "/*". Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-29 12:28:09 +08:00
/*
f2fs: add core inode operations This adds core functions to get, read, write, and evict an inode. Signed-off-by: Changman Lee <cm224.lee@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-02 16:10:40 +08:00
* Called at the last iput() if i_nlink is zero
*/
void f2fs_evict_inode(struct inode *inode)
{
struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
truncate_inode_pages(&inode->i_data, 0);
if (inode->i_ino == F2FS_NODE_INO(sbi) ||
inode->i_ino == F2FS_META_INO(sbi))
goto no_delete;
BUG_ON(atomic_read(&F2FS_I(inode)->dirty_dents));
remove_dirty_dir_inode(inode);
if (inode->i_nlink || is_bad_inode(inode))
goto no_delete;
set_inode_flag(F2FS_I(inode), FI_NO_ALLOC);
i_size_write(inode, 0);
if (F2FS_HAS_BLOCKS(inode))
f2fs_truncate(inode);
remove_inode_page(inode);
no_delete:
clear_inode(inode);
}