From aaf8c0b9ae042494cb4585883b15c1332de77840 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Wed, 26 Jun 2024 09:47:27 +0800 Subject: f2fs: reduce expensive checkpoint trigger frequency We may trigger high frequent checkpoint for below case: 1. mkdir /mnt/dir1; set dir1 encrypted 2. touch /mnt/file1; fsync /mnt/file1 3. mkdir /mnt/dir2; set dir2 encrypted 4. touch /mnt/file2; fsync /mnt/file2 ... Although, newly created dir and file are not related, due to commit bbf156f7afa7 ("f2fs: fix lost xattrs of directories"), we will trigger checkpoint whenever fsync() comes after a new encrypted dir created. In order to avoid such performance regression issue, let's record an entry including directory's ino in global cache whenever we update directory's xattr data, and then triggerring checkpoint() only if xattr metadata of target file's parent was updated. This patch updates to cover below no encryption case as well: 1) parent is checkpointed 2) set_xattr(dir) w/ new xnid 3) create(file) 4) fsync(file) Fixes: bbf156f7afa7 ("f2fs: fix lost xattrs of directories") Reported-by: wangzijie Reported-by: Zhiguo Niu Tested-by: Zhiguo Niu Reported-by: Yunlei He Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- include/trace/events/f2fs.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index ed794b5fefbe..2851c823095b 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -139,7 +139,8 @@ TRACE_DEFINE_ENUM(EX_BLOCK_AGE); { CP_NODE_NEED_CP, "node needs cp" }, \ { CP_FASTBOOT_MODE, "fastboot mode" }, \ { CP_SPEC_LOG_NUM, "log type is 2" }, \ - { CP_RECOVER_DIR, "dir needs recovery" }) + { CP_RECOVER_DIR, "dir needs recovery" }, \ + { CP_XATTR_DIR, "dir's xattr updated" }) #define show_shutdown_mode(type) \ __print_symbolic(type, \ -- cgit v1.2.3 From 8fb9f31984bdc0aaa1b0f7c7e7a27bd3137f8744 Mon Sep 17 00:00:00 2001 From: Zhiguo Niu Date: Thu, 1 Aug 2024 09:33:51 +0800 Subject: f2fs: clean up val{>>,<<}F2FS_BLKSIZE_BITS Use F2FS_BYTES_TO_BLK(bytes) and F2FS_BLK_TO_BYTES(blk) for cleanup Signed-off-by: Zhiguo Niu Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/checkpoint.c | 2 +- fs/f2fs/debug.c | 2 +- fs/f2fs/file.c | 6 +++--- fs/f2fs/node.c | 4 ++-- fs/f2fs/super.c | 2 +- include/linux/f2fs_fs.h | 2 +- 6 files changed, 9 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index bdd96329dddd..f3d22b8ef2ff 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -1551,7 +1551,7 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc) blk = start_blk + BLKS_PER_SEG(sbi) - nm_i->nat_bits_blocks; for (i = 0; i < nm_i->nat_bits_blocks; i++) f2fs_update_meta_page(sbi, nm_i->nat_bits + - (i << F2FS_BLKSIZE_BITS), blk + i); + F2FS_BLK_TO_BYTES(i), blk + i); } /* write out checkpoint buffer at block 0 */ diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c index 8b0e1e71b667..546b8ba91261 100644 --- a/fs/f2fs/debug.c +++ b/fs/f2fs/debug.c @@ -275,7 +275,7 @@ static void update_mem_info(struct f2fs_sb_info *sbi) /* build nm */ si->base_mem += sizeof(struct f2fs_nm_info); si->base_mem += __bitmap_size(sbi, NAT_BITMAP); - si->base_mem += (NM_I(sbi)->nat_bits_blocks << F2FS_BLKSIZE_BITS); + si->base_mem += F2FS_BLK_TO_BYTES(NM_I(sbi)->nat_bits_blocks); si->base_mem += NM_I(sbi)->nat_blocks * f2fs_bitmap_size(NAT_ENTRY_PER_BLOCK); si->base_mem += NM_I(sbi)->nat_blocks / 8; diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index a2d466ca2579..bddcb2cd945a 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -2979,9 +2979,9 @@ static int f2fs_move_file_range(struct file *file_in, loff_t pos_in, } f2fs_lock_op(sbi); - ret = __exchange_data_block(src, dst, pos_in >> F2FS_BLKSIZE_BITS, - pos_out >> F2FS_BLKSIZE_BITS, - len >> F2FS_BLKSIZE_BITS, false); + ret = __exchange_data_block(src, dst, F2FS_BYTES_TO_BLK(pos_in), + F2FS_BYTES_TO_BLK(pos_out), + F2FS_BYTES_TO_BLK(len), false); if (!ret) { if (dst_max_i_size) diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index 2fe182f73599..9e7a6e21f30c 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -3166,7 +3166,7 @@ static int __get_nat_bitmaps(struct f2fs_sb_info *sbi) nm_i->nat_bits_blocks = F2FS_BLK_ALIGN((nat_bits_bytes << 1) + 8); nm_i->nat_bits = f2fs_kvzalloc(sbi, - nm_i->nat_bits_blocks << F2FS_BLKSIZE_BITS, GFP_KERNEL); + F2FS_BLK_TO_BYTES(nm_i->nat_bits_blocks), GFP_KERNEL); if (!nm_i->nat_bits) return -ENOMEM; @@ -3185,7 +3185,7 @@ static int __get_nat_bitmaps(struct f2fs_sb_info *sbi) if (IS_ERR(page)) return PTR_ERR(page); - memcpy(nm_i->nat_bits + (i << F2FS_BLKSIZE_BITS), + memcpy(nm_i->nat_bits + F2FS_BLK_TO_BYTES(i), page_address(page), F2FS_BLKSIZE); f2fs_put_page(page, 1); } diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 4d2f3bb8e418..977f038f3557 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -3317,7 +3317,7 @@ loff_t max_file_blocks(struct inode *inode) * fit within U32_MAX + 1 data units. */ - result = min(result, (((loff_t)U32_MAX + 1) * 4096) >> F2FS_BLKSIZE_BITS); + result = min(result, F2FS_BYTES_TO_BLK(((loff_t)U32_MAX + 1) * 4096)); return result; } diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h index 01bee2b289c2..f17f89a259e2 100644 --- a/include/linux/f2fs_fs.h +++ b/include/linux/f2fs_fs.h @@ -19,7 +19,6 @@ #define F2FS_BLKSIZE_BITS PAGE_SHIFT /* bits for F2FS_BLKSIZE */ #define F2FS_MAX_EXTENSION 64 /* # of extension entries */ #define F2FS_EXTENSION_LEN 8 /* max size of extension */ -#define F2FS_BLK_ALIGN(x) (((x) + F2FS_BLKSIZE - 1) >> F2FS_BLKSIZE_BITS) #define NULL_ADDR ((block_t)0) /* used as block_t addresses */ #define NEW_ADDR ((block_t)-1) /* used as block_t addresses */ @@ -28,6 +27,7 @@ #define F2FS_BYTES_TO_BLK(bytes) ((bytes) >> F2FS_BLKSIZE_BITS) #define F2FS_BLK_TO_BYTES(blk) ((blk) << F2FS_BLKSIZE_BITS) #define F2FS_BLK_END_BYTES(blk) (F2FS_BLK_TO_BYTES(blk + 1) - 1) +#define F2FS_BLK_ALIGN(x) (F2FS_BYTES_TO_BLK((x) + F2FS_BLKSIZE - 1)) /* 0, 1(node nid), 2(meta nid) are reserved node id */ #define F2FS_RESERVED_NODE_NUM 3 -- cgit v1.2.3 From 884ee6dc85b959bc152f15bca80c30f06069e6c4 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Fri, 6 Sep 2024 14:27:24 +0800 Subject: f2fs: get rid of online repaire on corrupted directory syzbot reports a f2fs bug as below: kernel BUG at fs/f2fs/inode.c:896! RIP: 0010:f2fs_evict_inode+0x1598/0x15c0 fs/f2fs/inode.c:896 Call Trace: evict+0x532/0x950 fs/inode.c:704 dispose_list fs/inode.c:747 [inline] evict_inodes+0x5f9/0x690 fs/inode.c:797 generic_shutdown_super+0x9d/0x2d0 fs/super.c:627 kill_block_super+0x44/0x90 fs/super.c:1696 kill_f2fs_super+0x344/0x690 fs/f2fs/super.c:4898 deactivate_locked_super+0xc4/0x130 fs/super.c:473 cleanup_mnt+0x41f/0x4b0 fs/namespace.c:1373 task_work_run+0x24f/0x310 kernel/task_work.c:228 ptrace_notify+0x2d2/0x380 kernel/signal.c:2402 ptrace_report_syscall include/linux/ptrace.h:415 [inline] ptrace_report_syscall_exit include/linux/ptrace.h:477 [inline] syscall_exit_work+0xc6/0x190 kernel/entry/common.c:173 syscall_exit_to_user_mode_prepare kernel/entry/common.c:200 [inline] __syscall_exit_to_user_mode_work kernel/entry/common.c:205 [inline] syscall_exit_to_user_mode+0x279/0x370 kernel/entry/common.c:218 do_syscall_64+0x100/0x230 arch/x86/entry/common.c:89 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0010:f2fs_evict_inode+0x1598/0x15c0 fs/f2fs/inode.c:896 Online repaire on corrupted directory in f2fs_lookup() can generate dirty data/meta while racing w/ readonly remount, it may leave dirty inode after filesystem becomes readonly, however, checkpoint() will skips flushing dirty inode in a state of readonly mode, result in above panic. Let's get rid of online repaire in f2fs_lookup(), and leave the work to fsck.f2fs. Fixes: 510022a85839 ("f2fs: add F2FS_INLINE_DOTS to recover missing dot dentries") Reported-by: syzbot+ebea2790904673d7c618@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/000000000000a7b20f061ff2d56a@google.com Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 11 -------- fs/f2fs/namei.c | 68 ------------------------------------------------- include/linux/f2fs_fs.h | 2 +- 3 files changed, 1 insertion(+), 80 deletions(-) (limited to 'include') diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 8d850192dfdf..0abb7bb39296 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -790,7 +790,6 @@ enum { FI_NEED_IPU, /* used for ipu per file */ FI_ATOMIC_FILE, /* indicate atomic file */ FI_DATA_EXIST, /* indicate data exists */ - FI_INLINE_DOTS, /* indicate inline dot dentries */ FI_SKIP_WRITES, /* should skip data page writeback */ FI_OPU_WRITE, /* used for opu per file */ FI_DIRTY_FILE, /* indicate regular/symlink has dirty pages */ @@ -3065,7 +3064,6 @@ static inline void __mark_inode_dirty_flag(struct inode *inode, return; fallthrough; case FI_DATA_EXIST: - case FI_INLINE_DOTS: case FI_PIN_FILE: case FI_COMPRESS_RELEASED: f2fs_mark_inode_dirty_sync(inode, true); @@ -3189,8 +3187,6 @@ static inline void get_inline_info(struct inode *inode, struct f2fs_inode *ri) set_bit(FI_INLINE_DENTRY, fi->flags); if (ri->i_inline & F2FS_DATA_EXIST) set_bit(FI_DATA_EXIST, fi->flags); - if (ri->i_inline & F2FS_INLINE_DOTS) - set_bit(FI_INLINE_DOTS, fi->flags); if (ri->i_inline & F2FS_EXTRA_ATTR) set_bit(FI_EXTRA_ATTR, fi->flags); if (ri->i_inline & F2FS_PIN_FILE) @@ -3211,8 +3207,6 @@ static inline void set_raw_inline(struct inode *inode, struct f2fs_inode *ri) ri->i_inline |= F2FS_INLINE_DENTRY; if (is_inode_flag_set(inode, FI_DATA_EXIST)) ri->i_inline |= F2FS_DATA_EXIST; - if (is_inode_flag_set(inode, FI_INLINE_DOTS)) - ri->i_inline |= F2FS_INLINE_DOTS; if (is_inode_flag_set(inode, FI_EXTRA_ATTR)) ri->i_inline |= F2FS_EXTRA_ATTR; if (is_inode_flag_set(inode, FI_PIN_FILE)) @@ -3293,11 +3287,6 @@ static inline int f2fs_exist_data(struct inode *inode) return is_inode_flag_set(inode, FI_DATA_EXIST); } -static inline int f2fs_has_inline_dots(struct inode *inode) -{ - return is_inode_flag_set(inode, FI_INLINE_DOTS); -} - static inline int f2fs_is_mmap_file(struct inode *inode) { return is_inode_flag_set(inode, FI_MMAP_FILE); diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index 38b4750475db..57d46e1439de 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -457,62 +457,6 @@ struct dentry *f2fs_get_parent(struct dentry *child) return d_obtain_alias(f2fs_iget(child->d_sb, ino)); } -static int __recover_dot_dentries(struct inode *dir, nid_t pino) -{ - struct f2fs_sb_info *sbi = F2FS_I_SB(dir); - struct qstr dot = QSTR_INIT(".", 1); - struct f2fs_dir_entry *de; - struct page *page; - int err = 0; - - if (f2fs_readonly(sbi->sb)) { - f2fs_info(sbi, "skip recovering inline_dots inode (ino:%lu, pino:%u) in readonly mountpoint", - dir->i_ino, pino); - return 0; - } - - if (!S_ISDIR(dir->i_mode)) { - f2fs_err(sbi, "inconsistent inode status, skip recovering inline_dots inode (ino:%lu, i_mode:%u, pino:%u)", - dir->i_ino, dir->i_mode, pino); - set_sbi_flag(sbi, SBI_NEED_FSCK); - return -ENOTDIR; - } - - err = f2fs_dquot_initialize(dir); - if (err) - return err; - - f2fs_balance_fs(sbi, true); - - f2fs_lock_op(sbi); - - de = f2fs_find_entry(dir, &dot, &page); - if (de) { - f2fs_put_page(page, 0); - } else if (IS_ERR(page)) { - err = PTR_ERR(page); - goto out; - } else { - err = f2fs_do_add_link(dir, &dot, NULL, dir->i_ino, S_IFDIR); - if (err) - goto out; - } - - de = f2fs_find_entry(dir, &dotdot_name, &page); - if (de) - f2fs_put_page(page, 0); - else if (IS_ERR(page)) - err = PTR_ERR(page); - else - err = f2fs_do_add_link(dir, &dotdot_name, NULL, pino, S_IFDIR); -out: - if (!err) - clear_inode_flag(dir, FI_INLINE_DOTS); - - f2fs_unlock_op(sbi); - return err; -} - static struct dentry *f2fs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { @@ -522,7 +466,6 @@ static struct dentry *f2fs_lookup(struct inode *dir, struct dentry *dentry, struct dentry *new; nid_t ino = -1; int err = 0; - unsigned int root_ino = F2FS_ROOT_INO(F2FS_I_SB(dir)); struct f2fs_filename fname; trace_f2fs_lookup_start(dir, dentry, flags); @@ -558,17 +501,6 @@ static struct dentry *f2fs_lookup(struct inode *dir, struct dentry *dentry, goto out; } - if ((dir->i_ino == root_ino) && f2fs_has_inline_dots(dir)) { - err = __recover_dot_dentries(dir, root_ino); - if (err) - goto out_iput; - } - - if (f2fs_has_inline_dots(inode)) { - err = __recover_dot_dentries(inode, dir->i_ino); - if (err) - goto out_iput; - } if (IS_ENCRYPTED(dir) && (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) && !fscrypt_has_permitted_context(dir, inode)) { diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h index f17f89a259e2..b0b821edfd97 100644 --- a/include/linux/f2fs_fs.h +++ b/include/linux/f2fs_fs.h @@ -278,7 +278,7 @@ struct node_footer { #define F2FS_INLINE_DATA 0x02 /* file inline data flag */ #define F2FS_INLINE_DENTRY 0x04 /* file inline dentry flag */ #define F2FS_DATA_EXIST 0x08 /* file inline data exist flag */ -#define F2FS_INLINE_DOTS 0x10 /* file having implicit dot dentries */ +#define F2FS_INLINE_DOTS 0x10 /* file having implicit dot dentries (obsolete) */ #define F2FS_EXTRA_ATTR 0x20 /* file having extra attribute */ #define F2FS_PIN_FILE 0x40 /* file should not be gced */ #define F2FS_COMPRESS_RELEASED 0x80 /* file released compressed blocks */ -- cgit v1.2.3