From 66717260545b67b04ce6b004fff26de7141b2757 Mon Sep 17 00:00:00 2001 From: Elena Reshetova Date: Wed, 29 Nov 2017 13:19:31 +0200 Subject: posix_acl: convert posix_acl.a_refcount from atomic_t to refcount_t atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable posix_acl.a_refcount is used as pure reference counter. Convert it to refcount_t and fix up the operations. **Important note for maintainers: Some functions from refcount_t API defined in lib/refcount.c have different memory ordering guarantees than their atomic counterparts. The full comparison can be seen in https://lkml.org/lkml/2017/11/15/57 and it is hopefully soon in state to be merged to the documentation tree. Normally the differences should not matter since refcount_t provides enough guarantees to satisfy the refcounting use cases, but in some rare cases it might matter. Please double check that you don't have some undocumented memory guarantees for this variable usage. For the posix_acl.a_refcount it might make a difference in following places: - get_cached_acl(): increment in refcount_inc_not_zero() only guarantees control dependency on success vs. fully ordered atomic counterpart. However this operation is performed under rcu_read_lock(), so this should be fine. - posix_acl_release(): decrement in refcount_dec_and_test() only provides RELEASE ordering and control dependency on success vs. fully ordered atomic counterpart Suggested-by: Kees Cook Reviewed-by: David Windsor Reviewed-by: Hans Liljestrand Signed-off-by: Elena Reshetova Signed-off-by: Jaegeuk Kim --- include/linux/posix_acl.h | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/posix_acl.h b/include/linux/posix_acl.h index b2b7255ec7f5..540595a321a7 100644 --- a/include/linux/posix_acl.h +++ b/include/linux/posix_acl.h @@ -12,6 +12,7 @@ #include #include #include +#include #include struct posix_acl_entry { @@ -24,7 +25,7 @@ struct posix_acl_entry { }; struct posix_acl { - atomic_t a_refcount; + refcount_t a_refcount; struct rcu_head a_rcu; unsigned int a_count; struct posix_acl_entry a_entries[0]; @@ -41,7 +42,7 @@ static inline struct posix_acl * posix_acl_dup(struct posix_acl *acl) { if (acl) - atomic_inc(&acl->a_refcount); + refcount_inc(&acl->a_refcount); return acl; } @@ -51,7 +52,7 @@ posix_acl_dup(struct posix_acl *acl) static inline void posix_acl_release(struct posix_acl *acl) { - if (acl && atomic_dec_and_test(&acl->a_refcount)) + if (acl && refcount_dec_and_test(&acl->a_refcount)) kfree_rcu(acl, a_rcu); } -- cgit v1.2.3 From 0a007b97aad6e1700ef5c3815d14e88192cc1124 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Thu, 28 Dec 2017 08:09:44 -0800 Subject: f2fs: recover directory operations by fsync This fixes generic/342 which doesn't recover renamed file which was fsynced before. It will be done via another fsync on newly created file. Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/dir.c | 2 ++ fs/f2fs/f2fs.h | 2 ++ fs/f2fs/file.c | 3 +++ fs/f2fs/namei.c | 4 ++++ include/trace/events/f2fs.h | 3 ++- 5 files changed, 13 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c index 724304dc6143..f00b5ed8c011 100644 --- a/fs/f2fs/dir.c +++ b/fs/f2fs/dir.c @@ -713,6 +713,8 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page, f2fs_update_time(F2FS_I_SB(dir), REQ_TIME); + add_ino_entry(F2FS_I_SB(dir), dir->i_ino, TRANS_DIR_INO); + if (f2fs_has_inline_dentry(dir)) return f2fs_delete_inline_entry(dentry, page, dir, inode); diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index df00affe1450..46d3d89f3ac5 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -180,6 +180,7 @@ enum { ORPHAN_INO, /* for orphan ino list */ APPEND_INO, /* for append ino list */ UPDATE_INO, /* for update ino list */ + TRANS_DIR_INO, /* for trasactions dir ino list */ FLUSH_INO, /* for multiple device flushing */ MAX_INO_ENTRY, /* max. list */ }; @@ -924,6 +925,7 @@ enum cp_reason_type { CP_NODE_NEED_CP, CP_FASTBOOT_MODE, CP_SPEC_LOG_NUM, + CP_RECOVER_DIR, }; enum iostat_type { diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 1db68f1bcd77..322aab9d91b5 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -165,6 +165,9 @@ static inline enum cp_reason_type need_do_checkpoint(struct inode *inode) cp_reason = CP_FASTBOOT_MODE; else if (sbi->active_logs == 2) cp_reason = CP_SPEC_LOG_NUM; + else if (need_dentry_mark(sbi, inode->i_ino) && + exist_written_data(sbi, F2FS_I(inode)->i_pino, TRANS_DIR_INO)) + cp_reason = CP_RECOVER_DIR; return cp_reason; } diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index e3a0f4f65851..a885c6e659f8 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -935,6 +935,7 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry, } f2fs_i_links_write(old_dir, false); } + add_ino_entry(sbi, new_dir->i_ino, TRANS_DIR_INO); f2fs_unlock_op(sbi); @@ -1089,6 +1090,9 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry, } f2fs_mark_inode_dirty_sync(new_dir, false); + add_ino_entry(sbi, old_dir->i_ino, TRANS_DIR_INO); + add_ino_entry(sbi, new_dir->i_ino, TRANS_DIR_INO); + f2fs_unlock_op(sbi); if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir)) diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index 8f8dd42fa57b..06c87f9f720c 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -147,7 +147,8 @@ TRACE_DEFINE_ENUM(CP_TRIMMED); { CP_NO_SPC_ROLL, "no space roll forward" }, \ { CP_NODE_NEED_CP, "node needs cp" }, \ { CP_FASTBOOT_MODE, "fastboot mode" }, \ - { CP_SPEC_LOG_NUM, "log type is 2" }) + { CP_SPEC_LOG_NUM, "log type is 2" }, \ + { CP_RECOVER_DIR, "dir needs recovery" }) struct victim_sel_policy; struct f2fs_map_blocks; -- cgit v1.2.3 From 1ad71a27124caf0b68ddd3c92be01aa2b2a72b2a Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Thu, 7 Dec 2017 16:25:39 -0800 Subject: f2fs: add an ioctl to disable GC for specific file This patch gives a flag to disable GC on given file, which would be useful, when user wants to keep its block map. It also conducts in-place-update for dontmove file. Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 2 ++ fs/f2fs/f2fs.h | 29 ++++++++++++++++- fs/f2fs/file.c | 83 +++++++++++++++++++++++++++++++++++++++++++++++++ fs/f2fs/gc.c | 11 +++++++ fs/f2fs/gc.h | 2 ++ fs/f2fs/sysfs.c | 2 ++ include/linux/f2fs_fs.h | 9 +++++- 7 files changed, 136 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index cde2a3264b4c..945f15607484 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1479,6 +1479,8 @@ static inline bool need_inplace_update(struct f2fs_io_info *fio) { struct inode *inode = fio->page->mapping->host; + if (f2fs_is_pinned_file(inode)) + return true; if (S_ISDIR(inode->i_mode) || f2fs_is_atomic_file(inode)) return false; if (is_cold_data(fio->page)) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 1b2b0acbc83d..d5090b4d2633 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -356,6 +356,8 @@ static inline bool __has_cursum_space(struct f2fs_journal *journal, #define F2FS_IOC_GARBAGE_COLLECT_RANGE _IOW(F2FS_IOCTL_MAGIC, 11, \ struct f2fs_gc_range) #define F2FS_IOC_GET_FEATURES _IOR(F2FS_IOCTL_MAGIC, 12, __u32) +#define F2FS_IOC_SET_PIN_FILE _IOW(F2FS_IOCTL_MAGIC, 13, __u32) +#define F2FS_IOC_GET_PIN_FILE _IOR(F2FS_IOCTL_MAGIC, 14, __u32) #define F2FS_IOC_SET_ENCRYPTION_POLICY FS_IOC_SET_ENCRYPTION_POLICY #define F2FS_IOC_GET_ENCRYPTION_POLICY FS_IOC_GET_ENCRYPTION_POLICY @@ -593,7 +595,10 @@ struct f2fs_inode_info { unsigned long i_flags; /* keep an inode flags for ioctl */ unsigned char i_advise; /* use to give file attribute hints */ unsigned char i_dir_level; /* use for dentry level for large dir */ - unsigned int i_current_depth; /* use only in directory structure */ + union { + unsigned int i_current_depth; /* only for directory depth */ + unsigned short i_gc_failures; /* only for regular file */ + }; unsigned int i_pino; /* parent inode number */ umode_t i_acl_mode; /* keep file acl mode temporarily */ @@ -1142,6 +1147,9 @@ struct f2fs_sb_info { /* threshold for converting bg victims for fg */ u64 fggc_threshold; + /* threshold for gc trials on pinned files */ + u64 gc_pin_file_threshold; + /* maximum # of trials to find a victim segment for SSR and GC */ unsigned int max_victim_search; @@ -2141,6 +2149,7 @@ enum { FI_HOT_DATA, /* indicate file is hot */ FI_EXTRA_ATTR, /* indicate file has extra attribute */ FI_PROJ_INHERIT, /* indicate file inherits projectid */ + FI_PIN_FILE, /* indicate file should not be gced */ }; static inline void __mark_inode_dirty_flag(struct inode *inode, @@ -2155,6 +2164,7 @@ static inline void __mark_inode_dirty_flag(struct inode *inode, return; case FI_DATA_EXIST: case FI_INLINE_DOTS: + case FI_PIN_FILE: f2fs_mark_inode_dirty_sync(inode, true); } } @@ -2235,6 +2245,13 @@ static inline void f2fs_i_depth_write(struct inode *inode, unsigned int depth) f2fs_mark_inode_dirty_sync(inode, true); } +static inline void f2fs_i_gc_failures_write(struct inode *inode, + unsigned int count) +{ + F2FS_I(inode)->i_gc_failures = count; + f2fs_mark_inode_dirty_sync(inode, true); +} + static inline void f2fs_i_xnid_write(struct inode *inode, nid_t xnid) { F2FS_I(inode)->i_xattr_nid = xnid; @@ -2263,6 +2280,8 @@ static inline void get_inline_info(struct inode *inode, struct f2fs_inode *ri) 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) + set_bit(FI_PIN_FILE, &fi->flags); } static inline void set_raw_inline(struct inode *inode, struct f2fs_inode *ri) @@ -2281,6 +2300,8 @@ static inline void set_raw_inline(struct inode *inode, struct f2fs_inode *ri) 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)) + ri->i_inline |= F2FS_PIN_FILE; } static inline int f2fs_has_extra_attr(struct inode *inode) @@ -2326,6 +2347,11 @@ static inline int f2fs_has_inline_dots(struct inode *inode) return is_inode_flag_set(inode, FI_INLINE_DOTS); } +static inline bool f2fs_is_pinned_file(struct inode *inode) +{ + return is_inode_flag_set(inode, FI_PIN_FILE); +} + static inline bool f2fs_is_atomic_file(struct inode *inode) { return is_inode_flag_set(inode, FI_ATOMIC_FILE); @@ -2540,6 +2566,7 @@ int truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end); void truncate_data_blocks_range(struct dnode_of_data *dn, int count); long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg); +int f2fs_pin_file_control(struct inode *inode, bool inc); /* * inode.c diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 05779077cbf2..493457a200b1 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -2672,6 +2672,83 @@ static int f2fs_ioc_fssetxattr(struct file *filp, unsigned long arg) return 0; } +int f2fs_pin_file_control(struct inode *inode, bool inc) +{ + struct f2fs_inode_info *fi = F2FS_I(inode); + struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + + /* Use i_gc_failures for normal file as a risk signal. */ + if (inc) + f2fs_i_gc_failures_write(inode, fi->i_gc_failures + 1); + + if (fi->i_gc_failures > sbi->gc_pin_file_threshold) { + f2fs_msg(sbi->sb, KERN_WARNING, + "%s: Enable GC = ino %lx after %x GC trials\n", + __func__, inode->i_ino, fi->i_gc_failures); + clear_inode_flag(inode, FI_PIN_FILE); + return -EAGAIN; + } + return 0; +} + +static int f2fs_ioc_set_pin_file(struct file *filp, unsigned long arg) +{ + struct inode *inode = file_inode(filp); + __u32 pin; + int ret = 0; + + if (!inode_owner_or_capable(inode)) + return -EACCES; + + if (get_user(pin, (__u32 __user *)arg)) + return -EFAULT; + + if (!S_ISREG(inode->i_mode)) + return -EINVAL; + + if (f2fs_readonly(F2FS_I_SB(inode)->sb)) + return -EROFS; + + ret = mnt_want_write_file(filp); + if (ret) + return ret; + + inode_lock(inode); + + if (!pin) { + clear_inode_flag(inode, FI_PIN_FILE); + F2FS_I(inode)->i_gc_failures = 1; + goto done; + } + + if (f2fs_pin_file_control(inode, false)) { + ret = -EAGAIN; + goto out; + } + ret = f2fs_convert_inline_inode(inode); + if (ret) + goto out; + + set_inode_flag(inode, FI_PIN_FILE); + ret = F2FS_I(inode)->i_gc_failures; +done: + f2fs_update_time(F2FS_I_SB(inode), REQ_TIME); +out: + inode_unlock(inode); + mnt_drop_write_file(filp); + return ret; +} + +static int f2fs_ioc_get_pin_file(struct file *filp, unsigned long arg) +{ + struct inode *inode = file_inode(filp); + __u32 pin = 0; + + if (is_inode_flag_set(inode, FI_PIN_FILE)) + pin = F2FS_I(inode)->i_gc_failures; + return put_user(pin, (u32 __user *)arg); +} + long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { if (unlikely(f2fs_cp_error(F2FS_I_SB(file_inode(filp))))) @@ -2722,6 +2799,10 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return f2fs_ioc_fsgetxattr(filp, arg); case F2FS_IOC_FSSETXATTR: return f2fs_ioc_fssetxattr(filp, arg); + case F2FS_IOC_GET_PIN_FILE: + return f2fs_ioc_get_pin_file(filp, arg); + case F2FS_IOC_SET_PIN_FILE: + return f2fs_ioc_set_pin_file(filp, arg); default: return -ENOTTY; } @@ -2797,6 +2878,8 @@ long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case F2FS_IOC_GET_FEATURES: case F2FS_IOC_FSGETXATTR: case F2FS_IOC_FSSETXATTR: + case F2FS_IOC_GET_PIN_FILE: + case F2FS_IOC_SET_PIN_FILE: break; default: return -ENOIOCTLCMD; diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index d844dcb80570..33e79697e41c 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -624,6 +624,11 @@ static void move_data_block(struct inode *inode, block_t bidx, if (f2fs_is_atomic_file(inode)) goto out; + if (f2fs_is_pinned_file(inode)) { + f2fs_pin_file_control(inode, true); + goto out; + } + set_new_dnode(&dn, inode, NULL, NULL, 0); err = get_dnode_of_data(&dn, bidx, LOOKUP_NODE); if (err) @@ -720,6 +725,11 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type, if (f2fs_is_atomic_file(inode)) goto out; + if (f2fs_is_pinned_file(inode)) { + if (gc_type == FG_GC) + f2fs_pin_file_control(inode, true); + goto out; + } if (gc_type == BG_GC) { if (PageWriteback(page)) @@ -1091,6 +1101,7 @@ void build_gc_manager(struct f2fs_sb_info *sbi) sbi->fggc_threshold = div64_u64((main_count - ovp_count) * BLKS_PER_SEC(sbi), (main_count - resv_count)); + sbi->gc_pin_file_threshold = DEF_GC_FAILED_PINNED_FILES; /* give warm/cold data area from slower device */ if (sbi->s_ndevs && sbi->segs_per_sec == 1) diff --git a/fs/f2fs/gc.h b/fs/f2fs/gc.h index 9325191fab2d..b0045d4c8d1e 100644 --- a/fs/f2fs/gc.h +++ b/fs/f2fs/gc.h @@ -20,6 +20,8 @@ #define LIMIT_INVALID_BLOCK 40 /* percentage over total user space */ #define LIMIT_FREE_BLOCK 40 /* percentage over invalid + free space */ +#define DEF_GC_FAILED_PINNED_FILES 2048 + /* Search max. number of dirty segments to select a victim segment */ #define DEF_MAX_VICTIM_SEARCH 4096 /* covers 8GB */ diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index ab6028c332aa..41887e6ec1b3 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -301,6 +301,7 @@ F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, cp_interval, interval_time[CP_TIME]); F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, idle_interval, interval_time[REQ_TIME]); F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, iostat_enable, iostat_enable); F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, readdir_ra, readdir_ra); +F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, gc_pin_file_thresh, gc_pin_file_threshold); #ifdef CONFIG_F2FS_FAULT_INJECTION F2FS_RW_ATTR(FAULT_INFO_RATE, f2fs_fault_info, inject_rate, inject_rate); F2FS_RW_ATTR(FAULT_INFO_TYPE, f2fs_fault_info, inject_type, inject_type); @@ -349,6 +350,7 @@ static struct attribute *f2fs_attrs[] = { ATTR_LIST(idle_interval), ATTR_LIST(iostat_enable), ATTR_LIST(readdir_ra), + ATTR_LIST(gc_pin_file_thresh), #ifdef CONFIG_F2FS_FAULT_INJECTION ATTR_LIST(inject_rate), ATTR_LIST(inject_type), diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h index 43e98d30d2df..cfb522e6affc 100644 --- a/include/linux/f2fs_fs.h +++ b/include/linux/f2fs_fs.h @@ -212,6 +212,7 @@ struct f2fs_extent { #define F2FS_DATA_EXIST 0x08 /* file inline data exist flag */ #define F2FS_INLINE_DOTS 0x10 /* file having implicit dot dentries */ #define F2FS_EXTRA_ATTR 0x20 /* file having extra attribute */ +#define F2FS_PIN_FILE 0x40 /* file should not be gced */ struct f2fs_inode { __le16 i_mode; /* file mode */ @@ -229,7 +230,13 @@ struct f2fs_inode { __le32 i_ctime_nsec; /* change time in nano scale */ __le32 i_mtime_nsec; /* modification time in nano scale */ __le32 i_generation; /* file version (for NFS) */ - __le32 i_current_depth; /* only for directory depth */ + union { + __le32 i_current_depth; /* only for directory depth */ + __le16 i_gc_failures; /* + * # of gc failures on pinned file. + * only for regular files. + */ + }; __le32 i_xattr_nid; /* nid to save xattr */ __le32 i_flags; /* file attributes */ __le32 i_pino; /* parent inode number */ -- cgit v1.2.3 From f236792311f4575a1ca47240d3a74034096ef9e8 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Fri, 19 Jan 2018 13:42:33 -0800 Subject: f2fs: allow to recover node blocks given updated checkpoint If fsck.f2fs changes crc, we have no way to recover some inode blocks by roll- forward recovery. Let's relax the condition to recover them. Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/checkpoint.c | 1 + fs/f2fs/node.h | 4 ++++ include/linux/f2fs_fs.h | 1 + 3 files changed, 6 insertions(+) (limited to 'include') diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index 9c7596f7daae..512dca8abc7d 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -1157,6 +1157,7 @@ static void update_ckpt_flags(struct f2fs_sb_info *sbi, struct cp_control *cpc) /* set this flag to activate crc|cp_ver for recovery */ __set_ckpt_flags(ckpt, CP_CRC_RECOVERY_FLAG); + __clear_ckpt_flags(ckpt, CP_NOCRC_RECOVERY_FLAG); spin_unlock_irqrestore(&sbi->cp_lock, flags); } diff --git a/fs/f2fs/node.h b/fs/f2fs/node.h index 0ee3e5ff49a3..081ef0d672bf 100644 --- a/fs/f2fs/node.h +++ b/fs/f2fs/node.h @@ -305,6 +305,10 @@ static inline bool is_recoverable_dnode(struct page *page) struct f2fs_checkpoint *ckpt = F2FS_CKPT(F2FS_P_SB(page)); __u64 cp_ver = cur_cp_version(ckpt); + /* Don't care crc part, if fsck.f2fs sets it. */ + if (__is_set_ckpt_flags(ckpt, CP_NOCRC_RECOVERY_FLAG)) + return (cp_ver << 32) == (cpver_of_node(page) << 32); + if (__is_set_ckpt_flags(ckpt, CP_CRC_RECOVERY_FLAG)) cp_ver |= (cur_cp_crc(ckpt) << 32); diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h index cfb522e6affc..6eed677b6d9a 100644 --- a/include/linux/f2fs_fs.h +++ b/include/linux/f2fs_fs.h @@ -117,6 +117,7 @@ struct f2fs_super_block { /* * For checkpoint */ +#define CP_NOCRC_RECOVERY_FLAG 0x00000200 #define CP_TRIMMED_FLAG 0x00000100 #define CP_NAT_BITS_FLAG 0x00000080 #define CP_CRC_RECOVERY_FLAG 0x00000040 -- cgit v1.2.3 From 1c1d35df71104c76a4a2e25862926f22c334c9d6 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Thu, 25 Jan 2018 14:54:42 +0800 Subject: f2fs: support inode creation time This patch adds creation time field in inode layout to support showing kstat.btime in ->statx. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 7 +++++++ fs/f2fs/file.c | 9 +++++++++ fs/f2fs/inode.c | 15 +++++++++++++++ fs/f2fs/namei.c | 3 ++- fs/f2fs/sysfs.c | 7 +++++++ include/linux/f2fs_fs.h | 4 +++- 6 files changed, 43 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index b7ba496af28f..6300ac5bcbe4 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -124,6 +124,7 @@ struct f2fs_mount_info { #define F2FS_FEATURE_INODE_CHKSUM 0x0020 #define F2FS_FEATURE_FLEXIBLE_INLINE_XATTR 0x0040 #define F2FS_FEATURE_QUOTA_INO 0x0080 +#define F2FS_FEATURE_INODE_CRTIME 0x0100 #define F2FS_HAS_FEATURE(sb, mask) \ ((F2FS_SB(sb)->raw_super->feature & cpu_to_le32(mask)) != 0) @@ -635,6 +636,7 @@ struct f2fs_inode_info { int i_extra_isize; /* size of extra space located in i_addr */ kprojid_t i_projid; /* id for project quota */ int i_inline_xattr_size; /* inline xattr size */ + struct timespec i_crtime; /* inode creation time */ }; static inline void get_extent_info(struct extent_info *ext, @@ -3205,6 +3207,11 @@ static inline int f2fs_sb_has_quota_ino(struct super_block *sb) return F2FS_HAS_FEATURE(sb, F2FS_FEATURE_QUOTA_INO); } +static inline int f2fs_sb_has_inode_crtime(struct super_block *sb) +{ + return F2FS_HAS_FEATURE(sb, F2FS_FEATURE_INODE_CRTIME); +} + #ifdef CONFIG_BLK_DEV_ZONED static inline int get_blkz_type(struct f2fs_sb_info *sbi, struct block_device *bdev, block_t blkaddr) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 84306c718e68..672a542e5464 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -672,8 +672,17 @@ int f2fs_getattr(const struct path *path, struct kstat *stat, { struct inode *inode = d_inode(path->dentry); struct f2fs_inode_info *fi = F2FS_I(inode); + struct f2fs_inode *ri; unsigned int flags; + if (f2fs_has_extra_attr(inode) && + f2fs_sb_has_inode_crtime(inode->i_sb) && + F2FS_FITS_IN_INODE(ri, fi->i_extra_isize, i_crtime)) { + stat->result_mask |= STATX_BTIME; + stat->btime.tv_sec = fi->i_crtime.tv_sec; + stat->btime.tv_nsec = fi->i_crtime.tv_nsec; + } + flags = fi->i_flags & (FS_FL_USER_VISIBLE | FS_PROJINHERIT_FL); if (flags & FS_APPEND_FL) stat->attributes |= STATX_ATTR_APPEND; diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index 1dc77a40d0ad..89c838bfb067 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -278,6 +278,12 @@ static int do_read_inode(struct inode *inode) i_projid = F2FS_DEF_PROJID; fi->i_projid = make_kprojid(&init_user_ns, i_projid); + if (f2fs_has_extra_attr(inode) && f2fs_sb_has_inode_crtime(sbi->sb) && + F2FS_FITS_IN_INODE(ri, fi->i_extra_isize, i_crtime)) { + fi->i_crtime.tv_sec = le64_to_cpu(ri->i_crtime); + fi->i_crtime.tv_nsec = le32_to_cpu(ri->i_crtime_nsec); + } + f2fs_put_page(node_page, 1); stat_inc_inline_xattr(inode); @@ -421,6 +427,15 @@ void update_inode(struct inode *inode, struct page *node_page) F2FS_I(inode)->i_projid); ri->i_projid = cpu_to_le32(i_projid); } + + if (f2fs_sb_has_inode_crtime(F2FS_I_SB(inode)->sb) && + F2FS_FITS_IN_INODE(ri, F2FS_I(inode)->i_extra_isize, + i_crtime)) { + ri->i_crtime = + cpu_to_le64(F2FS_I(inode)->i_crtime.tv_sec); + ri->i_crtime_nsec = + cpu_to_le32(F2FS_I(inode)->i_crtime.tv_nsec); + } } __set_inode_rdev(inode, ri); diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index 3ee97ba9d2d7..c4c94c7e9f4f 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -50,7 +50,8 @@ static struct inode *f2fs_new_inode(struct inode *dir, umode_t mode) inode->i_ino = ino; inode->i_blocks = 0; - inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); + inode->i_mtime = inode->i_atime = inode->i_ctime = + F2FS_I(inode)->i_crtime = current_time(inode); inode->i_generation = sbi->s_next_generation++; err = insert_inode_locked(inode); diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index 41887e6ec1b3..d978c7b6ea04 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -113,6 +113,9 @@ static ssize_t features_show(struct f2fs_attr *a, if (f2fs_sb_has_quota_ino(sb)) len += snprintf(buf + len, PAGE_SIZE - len, "%s%s", len ? ", " : "", "quota_ino"); + if (f2fs_sb_has_inode_crtime(sb)) + len += snprintf(buf + len, PAGE_SIZE - len, "%s%s", + len ? ", " : "", "inode_crtime"); len += snprintf(buf + len, PAGE_SIZE - len, "\n"); return len; } @@ -232,6 +235,7 @@ enum feat_id { FEAT_INODE_CHECKSUM, FEAT_FLEXIBLE_INLINE_XATTR, FEAT_QUOTA_INO, + FEAT_INODE_CRTIME, }; static ssize_t f2fs_feature_show(struct f2fs_attr *a, @@ -246,6 +250,7 @@ static ssize_t f2fs_feature_show(struct f2fs_attr *a, case FEAT_INODE_CHECKSUM: case FEAT_FLEXIBLE_INLINE_XATTR: case FEAT_QUOTA_INO: + case FEAT_INODE_CRTIME: return snprintf(buf, PAGE_SIZE, "supported\n"); } return 0; @@ -323,6 +328,7 @@ F2FS_FEATURE_RO_ATTR(project_quota, FEAT_PROJECT_QUOTA); F2FS_FEATURE_RO_ATTR(inode_checksum, FEAT_INODE_CHECKSUM); F2FS_FEATURE_RO_ATTR(flexible_inline_xattr, FEAT_FLEXIBLE_INLINE_XATTR); F2FS_FEATURE_RO_ATTR(quota_ino, FEAT_QUOTA_INO); +F2FS_FEATURE_RO_ATTR(inode_crtime, FEAT_INODE_CRTIME); #define ATTR_LIST(name) (&f2fs_attr_##name.attr) static struct attribute *f2fs_attrs[] = { @@ -376,6 +382,7 @@ static struct attribute *f2fs_feat_attrs[] = { ATTR_LIST(inode_checksum), ATTR_LIST(flexible_inline_xattr), ATTR_LIST(quota_ino), + ATTR_LIST(inode_crtime), NULL, }; diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h index 6eed677b6d9a..58aecb60ea51 100644 --- a/include/linux/f2fs_fs.h +++ b/include/linux/f2fs_fs.h @@ -253,8 +253,10 @@ struct f2fs_inode { __le16 i_inline_xattr_size; /* inline xattr size, unit: 4 bytes */ __le32 i_projid; /* project id */ __le32 i_inode_checksum;/* inode meta checksum */ + __le64 i_crtime; /* creation time */ + __le32 i_crtime_nsec; /* creation time in nano scale */ __le32 i_extra_end[0]; /* for attribute size calculation */ - }; + } __packed; __le32 i_addr[DEF_ADDRS_PER_INODE]; /* Pointers to data blocks */ }; __le32 i_nid[DEF_NIDS_PER_INODE]; /* direct(2), indirect(2), -- cgit v1.2.3