diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2022-05-31 16:52:59 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2022-05-31 16:52:59 -0700 |
commit | 1501f707d2b24316b41d45bdc95a73bc8cc8dd49 (patch) | |
tree | 432a2c255fd269f4910d6c36cfe9eb58cdc92f5d /fs/f2fs/file.c | |
parent | 2a5699b0de4ee623d77f183c8e8e62691bd60a70 (diff) | |
parent | 2d1fe8a86bf5e0663866fd0da83c2af1e1b0e362 (diff) | |
download | linux-1501f707d2b24316b41d45bdc95a73bc8cc8dd49.tar.bz2 |
Merge tag 'f2fs-for-5.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs
Pull f2fs updates from Jaegeuk Kim:
"In this round, we've refactored the existing atomic write support
implemented by in-memory operations to have storing data in disk
temporarily, which can give us a benefit to accept more atomic writes.
At the same time, we removed the existing volatile write support.
We've also revisited the file pinning and GC flows and found some
corner cases which contributeed abnormal system behaviours.
As usual, there're several minor code refactoring for readability,
sanity check, and clean ups.
Enhancements:
- allow compression for mmap files in compress_mode=user
- kill volatile write support
- change the current atomic write way
- give priority to select unpinned section for foreground GC
- introduce data read/write showing path info
- remove unnecessary f2fs_lock_op in f2fs_new_inode
Bug fixes:
- fix the file pinning flow during checkpoint=disable and GCs
- fix foreground and background GCs to select the right victims and
get free sections on time
- fix GC flags on defragmenting pages
- avoid an infinite loop to flush node pages
- fix fallocate to use file_modified to update permissions
consistently"
* tag 'f2fs-for-5.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (40 commits)
f2fs: fix to tag gcing flag on page during file defragment
f2fs: replace F2FS_I(inode) and sbi by the local variable
f2fs: add f2fs_init_write_merge_io function
f2fs: avoid unneeded error handling for revoke_entry_slab allocation
f2fs: allow compression for mmap files in compress_mode=user
f2fs: fix typo in comment
f2fs: make f2fs_read_inline_data() more readable
f2fs: fix to do sanity check for inline inode
f2fs: fix fallocate to use file_modified to update permissions consistently
f2fs: don't use casefolded comparison for "." and ".."
f2fs: do not stop GC when requiring a free section
f2fs: keep wait_ms if EAGAIN happens
f2fs: introduce f2fs_gc_control to consolidate f2fs_gc parameters
f2fs: reject test_dummy_encryption when !CONFIG_FS_ENCRYPTION
f2fs: kill volatile write support
f2fs: change the current atomic write way
f2fs: don't need inode lock for system hidden quota
f2fs: stop allocating pinned sections if EAGAIN happens
f2fs: skip GC if possible when checkpoint disabling
f2fs: give priority to select unpinned section for foreground GC
...
Diffstat (limited to 'fs/f2fs/file.c')
-rw-r--r-- | fs/f2fs/file.c | 307 |
1 files changed, 137 insertions, 170 deletions
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 100637b1adb3..bd14cef1b08f 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -372,7 +372,8 @@ sync_nodes: f2fs_remove_ino_entry(sbi, ino, APPEND_INO); clear_inode_flag(inode, FI_APPEND_WRITE); flush_out: - if (!atomic && F2FS_OPTION(sbi).fsync_mode != FSYNC_MODE_NOBARRIER) + if ((!atomic && F2FS_OPTION(sbi).fsync_mode != FSYNC_MODE_NOBARRIER) || + (atomic && !test_opt(sbi, NOBARRIER) && f2fs_sb_has_blkzoned(sbi))) ret = f2fs_issue_flush(sbi, inode->i_ino); if (!ret) { f2fs_remove_ino_entry(sbi, ino, UPDATE_INO); @@ -1437,11 +1438,19 @@ static int f2fs_do_zero_range(struct dnode_of_data *dn, pgoff_t start, ret = -ENOSPC; break; } - if (dn->data_blkaddr != NEW_ADDR) { - f2fs_invalidate_blocks(sbi, dn->data_blkaddr); - dn->data_blkaddr = NEW_ADDR; - f2fs_set_data_blkaddr(dn); + + if (dn->data_blkaddr == NEW_ADDR) + continue; + + if (!f2fs_is_valid_blkaddr(sbi, dn->data_blkaddr, + DATA_GENERIC_ENHANCE)) { + ret = -EFSCORRUPTED; + break; } + + f2fs_invalidate_blocks(sbi, dn->data_blkaddr); + dn->data_blkaddr = NEW_ADDR; + f2fs_set_data_blkaddr(dn); } f2fs_update_extent_cache_range(dn, start, 0, index - start); @@ -1638,6 +1647,11 @@ static int expand_inode_data(struct inode *inode, loff_t offset, struct f2fs_map_blocks map = { .m_next_pgofs = NULL, .m_next_extent = NULL, .m_seg_type = NO_CHECK_TYPE, .m_may_create = true }; + struct f2fs_gc_control gc_control = { .victim_segno = NULL_SEGNO, + .init_gc_type = FG_GC, + .should_migrate_blocks = false, + .err_gc_skipped = true, + .nr_free_secs = 0 }; pgoff_t pg_start, pg_end; loff_t new_size = i_size_read(inode); loff_t off_end; @@ -1675,8 +1689,8 @@ next_alloc: if (has_not_enough_free_secs(sbi, 0, GET_SEC_FROM_SEG(sbi, overprovision_segments(sbi)))) { f2fs_down_write(&sbi->gc_lock); - err = f2fs_gc(sbi, true, false, false, NULL_SEGNO); - if (err && err != -ENODATA && err != -EAGAIN) + err = f2fs_gc(sbi, &gc_control); + if (err && err != -ENODATA) goto out_err; } @@ -1766,6 +1780,10 @@ static long f2fs_fallocate(struct file *file, int mode, inode_lock(inode); + ret = file_modified(file); + if (ret) + goto out; + if (mode & FALLOC_FL_PUNCH_HOLE) { if (offset >= inode->i_size) goto out; @@ -1804,16 +1822,8 @@ static int f2fs_release_file(struct inode *inode, struct file *filp) atomic_read(&inode->i_writecount) != 1) return 0; - /* some remained atomic pages should discarded */ if (f2fs_is_atomic_file(inode)) - f2fs_drop_inmem_pages(inode); - if (f2fs_is_volatile_file(inode)) { - set_inode_flag(inode, FI_DROP_CACHE); - filemap_fdatawrite(inode->i_mapping); - clear_inode_flag(inode, FI_DROP_CACHE); - clear_inode_flag(inode, FI_VOLATILE_FILE); - stat_dec_volatile_write(inode); - } + f2fs_abort_atomic_write(inode, true); return 0; } @@ -1828,8 +1838,8 @@ static int f2fs_file_flush(struct file *file, fl_owner_t id) * before dropping file lock, it needs to do in ->flush. */ if (f2fs_is_atomic_file(inode) && - F2FS_I(inode)->inmem_task == current) - f2fs_drop_inmem_pages(inode); + F2FS_I(inode)->atomic_write_task == current) + f2fs_abort_atomic_write(inode, true); return 0; } @@ -1992,6 +2002,7 @@ static int f2fs_ioc_start_atomic_write(struct file *filp) struct user_namespace *mnt_userns = file_mnt_user_ns(filp); struct f2fs_inode_info *fi = F2FS_I(inode); struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + struct inode *pinode; int ret; if (!inode_owner_or_capable(mnt_userns, inode)) @@ -2014,44 +2025,55 @@ static int f2fs_ioc_start_atomic_write(struct file *filp) goto out; } - if (f2fs_is_atomic_file(inode)) { - if (is_inode_flag_set(inode, FI_ATOMIC_REVOKE_REQUEST)) - ret = -EINVAL; + if (f2fs_is_atomic_file(inode)) goto out; - } ret = f2fs_convert_inline_inode(inode); if (ret) goto out; - f2fs_down_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); + f2fs_down_write(&fi->i_gc_rwsem[WRITE]); /* * Should wait end_io to count F2FS_WB_CP_DATA correctly by * f2fs_is_atomic_file. */ if (get_dirty_pages(inode)) - f2fs_warn(F2FS_I_SB(inode), "Unexpected flush for atomic writes: ino=%lu, npages=%u", + f2fs_warn(sbi, "Unexpected flush for atomic writes: ino=%lu, npages=%u", inode->i_ino, get_dirty_pages(inode)); ret = filemap_write_and_wait_range(inode->i_mapping, 0, LLONG_MAX); if (ret) { - f2fs_up_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); + f2fs_up_write(&fi->i_gc_rwsem[WRITE]); + goto out; + } + + /* Create a COW inode for atomic write */ + pinode = f2fs_iget(inode->i_sb, fi->i_pino); + if (IS_ERR(pinode)) { + f2fs_up_write(&fi->i_gc_rwsem[WRITE]); + ret = PTR_ERR(pinode); + goto out; + } + + ret = f2fs_get_tmpfile(mnt_userns, pinode, &fi->cow_inode); + iput(pinode); + if (ret) { + f2fs_up_write(&fi->i_gc_rwsem[WRITE]); goto out; } + f2fs_i_size_write(fi->cow_inode, i_size_read(inode)); spin_lock(&sbi->inode_lock[ATOMIC_FILE]); - if (list_empty(&fi->inmem_ilist)) - list_add_tail(&fi->inmem_ilist, &sbi->inode_list[ATOMIC_FILE]); sbi->atomic_files++; spin_unlock(&sbi->inode_lock[ATOMIC_FILE]); - /* add inode in inmem_list first and set atomic_file */ set_inode_flag(inode, FI_ATOMIC_FILE); - clear_inode_flag(inode, FI_ATOMIC_REVOKE_REQUEST); - f2fs_up_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); + set_inode_flag(fi->cow_inode, FI_ATOMIC_FILE); + clear_inode_flag(fi->cow_inode, FI_INLINE_DATA); + f2fs_up_write(&fi->i_gc_rwsem[WRITE]); - f2fs_update_time(F2FS_I_SB(inode), REQ_TIME); - F2FS_I(inode)->inmem_task = current; + f2fs_update_time(sbi, REQ_TIME); + fi->atomic_write_task = current; stat_update_max_atomic_write(inode); out: inode_unlock(inode); @@ -2076,127 +2098,20 @@ static int f2fs_ioc_commit_atomic_write(struct file *filp) inode_lock(inode); - if (f2fs_is_volatile_file(inode)) { - ret = -EINVAL; - goto err_out; - } - if (f2fs_is_atomic_file(inode)) { - ret = f2fs_commit_inmem_pages(inode); + ret = f2fs_commit_atomic_write(inode); if (ret) - goto err_out; + goto unlock_out; ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 0, true); if (!ret) - f2fs_drop_inmem_pages(inode); + f2fs_abort_atomic_write(inode, false); } else { ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 1, false); } -err_out: - if (is_inode_flag_set(inode, FI_ATOMIC_REVOKE_REQUEST)) { - clear_inode_flag(inode, FI_ATOMIC_REVOKE_REQUEST); - ret = -EINVAL; - } - inode_unlock(inode); - mnt_drop_write_file(filp); - return ret; -} - -static int f2fs_ioc_start_volatile_write(struct file *filp) -{ - struct inode *inode = file_inode(filp); - struct user_namespace *mnt_userns = file_mnt_user_ns(filp); - int ret; - - if (!inode_owner_or_capable(mnt_userns, inode)) - return -EACCES; - - if (!S_ISREG(inode->i_mode)) - return -EINVAL; - - ret = mnt_want_write_file(filp); - if (ret) - return ret; - - inode_lock(inode); - - if (f2fs_is_volatile_file(inode)) - goto out; - - ret = f2fs_convert_inline_inode(inode); - if (ret) - goto out; - - stat_inc_volatile_write(inode); - stat_update_max_volatile_write(inode); - - set_inode_flag(inode, FI_VOLATILE_FILE); - f2fs_update_time(F2FS_I_SB(inode), REQ_TIME); -out: - inode_unlock(inode); - mnt_drop_write_file(filp); - return ret; -} - -static int f2fs_ioc_release_volatile_write(struct file *filp) -{ - struct inode *inode = file_inode(filp); - struct user_namespace *mnt_userns = file_mnt_user_ns(filp); - int ret; - - if (!inode_owner_or_capable(mnt_userns, inode)) - return -EACCES; - - ret = mnt_want_write_file(filp); - if (ret) - return ret; - - inode_lock(inode); - - if (!f2fs_is_volatile_file(inode)) - goto out; - - if (!f2fs_is_first_block_written(inode)) { - ret = truncate_partial_data_page(inode, 0, true); - goto out; - } - - ret = punch_hole(inode, 0, F2FS_BLKSIZE); -out: - inode_unlock(inode); - mnt_drop_write_file(filp); - return ret; -} - -static int f2fs_ioc_abort_volatile_write(struct file *filp) -{ - struct inode *inode = file_inode(filp); - struct user_namespace *mnt_userns = file_mnt_user_ns(filp); - int ret; - - if (!inode_owner_or_capable(mnt_userns, inode)) - return -EACCES; - - ret = mnt_want_write_file(filp); - if (ret) - return ret; - - inode_lock(inode); - - if (f2fs_is_atomic_file(inode)) - f2fs_drop_inmem_pages(inode); - if (f2fs_is_volatile_file(inode)) { - clear_inode_flag(inode, FI_VOLATILE_FILE); - stat_dec_volatile_write(inode); - ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 0, true); - } - - clear_inode_flag(inode, FI_ATOMIC_REVOKE_REQUEST); - +unlock_out: inode_unlock(inode); - mnt_drop_write_file(filp); - f2fs_update_time(F2FS_I_SB(inode), REQ_TIME); return ret; } @@ -2437,6 +2352,10 @@ static int f2fs_ioc_gc(struct file *filp, unsigned long arg) { struct inode *inode = file_inode(filp); struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + struct f2fs_gc_control gc_control = { .victim_segno = NULL_SEGNO, + .no_bg_gc = false, + .should_migrate_blocks = false, + .nr_free_secs = 0 }; __u32 sync; int ret; @@ -2462,7 +2381,9 @@ static int f2fs_ioc_gc(struct file *filp, unsigned long arg) f2fs_down_write(&sbi->gc_lock); } - ret = f2fs_gc(sbi, sync, true, false, NULL_SEGNO); + gc_control.init_gc_type = sync ? FG_GC : BG_GC; + gc_control.err_gc_skipped = sync; + ret = f2fs_gc(sbi, &gc_control); out: mnt_drop_write_file(filp); return ret; @@ -2471,6 +2392,12 @@ out: static int __f2fs_ioc_gc_range(struct file *filp, struct f2fs_gc_range *range) { struct f2fs_sb_info *sbi = F2FS_I_SB(file_inode(filp)); + struct f2fs_gc_control gc_control = { + .init_gc_type = range->sync ? FG_GC : BG_GC, + .no_bg_gc = false, + .should_migrate_blocks = false, + .err_gc_skipped = range->sync, + .nr_free_secs = 0 }; u64 end; int ret; @@ -2498,8 +2425,8 @@ do_more: f2fs_down_write(&sbi->gc_lock); } - ret = f2fs_gc(sbi, range->sync, true, false, - GET_SEGNO(sbi, range->start)); + gc_control.victim_segno = GET_SEGNO(sbi, range->start); + ret = f2fs_gc(sbi, &gc_control); if (ret) { if (ret == -EBUSY) ret = -EAGAIN; @@ -2674,6 +2601,7 @@ do_map: } set_page_dirty(page); + set_page_private_gcing(page); f2fs_put_page(page, 1); idx++; @@ -2913,6 +2841,11 @@ static int f2fs_ioc_flush_device(struct file *filp, unsigned long arg) unsigned int start_segno = 0, end_segno = 0; unsigned int dev_start_segno = 0, dev_end_segno = 0; struct f2fs_flush_device range; + struct f2fs_gc_control gc_control = { + .init_gc_type = FG_GC, + .should_migrate_blocks = true, + .err_gc_skipped = true, + .nr_free_secs = 0 }; int ret; if (!capable(CAP_SYS_ADMIN)) @@ -2956,7 +2889,9 @@ static int f2fs_ioc_flush_device(struct file *filp, unsigned long arg) sm->last_victim[GC_CB] = end_segno + 1; sm->last_victim[GC_GREEDY] = end_segno + 1; sm->last_victim[ALLOC_NEXT] = end_segno + 1; - ret = f2fs_gc(sbi, true, true, true, start_segno); + + gc_control.victim_segno = start_segno; + ret = f2fs_gc(sbi, &gc_control); if (ret == -EAGAIN) ret = 0; else if (ret < 0) @@ -3017,7 +2952,7 @@ static int f2fs_ioc_setproject(struct inode *inode, __u32 projid) kprojid = make_kprojid(&init_user_ns, (projid_t)projid); - if (projid_eq(kprojid, F2FS_I(inode)->i_projid)) + if (projid_eq(kprojid, fi->i_projid)) return 0; err = -EPERM; @@ -3037,7 +2972,7 @@ static int f2fs_ioc_setproject(struct inode *inode, __u32 projid) if (err) goto out_unlock; - F2FS_I(inode)->i_projid = kprojid; + fi->i_projid = kprojid; inode->i_ctime = current_time(inode); f2fs_mark_inode_dirty_sync(inode, true); out_unlock: @@ -3987,7 +3922,7 @@ static int f2fs_ioc_decompress_file(struct file *filp, unsigned long arg) struct f2fs_inode_info *fi = F2FS_I(inode); pgoff_t page_idx = 0, last_idx; unsigned int blk_per_seg = sbi->blocks_per_seg; - int cluster_size = F2FS_I(inode)->i_cluster_size; + int cluster_size = fi->i_cluster_size; int count, ret; if (!f2fs_sb_has_compression(sbi) || @@ -4010,11 +3945,6 @@ static int f2fs_ioc_decompress_file(struct file *filp, unsigned long arg) goto out; } - if (f2fs_is_mmap_file(inode)) { - ret = -EBUSY; - goto out; - } - ret = filemap_write_and_wait_range(inode->i_mapping, 0, LLONG_MAX); if (ret) goto out; @@ -4082,11 +4012,6 @@ static int f2fs_ioc_compress_file(struct file *filp, unsigned long arg) goto out; } - if (f2fs_is_mmap_file(inode)) { - ret = -EBUSY; - goto out; - } - ret = filemap_write_and_wait_range(inode->i_mapping, 0, LLONG_MAX); if (ret) goto out; @@ -4136,11 +4061,9 @@ static long __f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) case F2FS_IOC_COMMIT_ATOMIC_WRITE: return f2fs_ioc_commit_atomic_write(filp); case F2FS_IOC_START_VOLATILE_WRITE: - return f2fs_ioc_start_volatile_write(filp); case F2FS_IOC_RELEASE_VOLATILE_WRITE: - return f2fs_ioc_release_volatile_write(filp); case F2FS_IOC_ABORT_VOLATILE_WRITE: - return f2fs_ioc_abort_volatile_write(filp); + return -EOPNOTSUPP; case F2FS_IOC_SHUTDOWN: return f2fs_ioc_shutdown(filp, arg); case FITRIM: @@ -4328,17 +4251,39 @@ out: static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) { struct inode *inode = file_inode(iocb->ki_filp); + const loff_t pos = iocb->ki_pos; ssize_t ret; if (!f2fs_is_compress_backend_ready(inode)) return -EOPNOTSUPP; - if (f2fs_should_use_dio(inode, iocb, to)) - return f2fs_dio_read_iter(iocb, to); + if (trace_f2fs_dataread_start_enabled()) { + char *p = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); + char *path; + + if (!p) + goto skip_read_trace; + + path = dentry_path_raw(file_dentry(iocb->ki_filp), p, PATH_MAX); + if (IS_ERR(path)) { + kfree(p); + goto skip_read_trace; + } - ret = filemap_read(iocb, to, 0); - if (ret > 0) - f2fs_update_iostat(F2FS_I_SB(inode), APP_BUFFERED_READ_IO, ret); + trace_f2fs_dataread_start(inode, pos, iov_iter_count(to), + current->pid, path, current->comm); + kfree(p); + } +skip_read_trace: + if (f2fs_should_use_dio(inode, iocb, to)) { + ret = f2fs_dio_read_iter(iocb, to); + } else { + ret = filemap_read(iocb, to, 0); + if (ret > 0) + f2fs_update_iostat(F2FS_I_SB(inode), APP_BUFFERED_READ_IO, ret); + } + if (trace_f2fs_dataread_end_enabled()) + trace_f2fs_dataread_end(inode, pos, ret); return ret; } @@ -4630,14 +4575,36 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) /* Possibly preallocate the blocks for the write. */ target_size = iocb->ki_pos + iov_iter_count(from); preallocated = f2fs_preallocate_blocks(iocb, from, dio); - if (preallocated < 0) + if (preallocated < 0) { ret = preallocated; - else + } else { + if (trace_f2fs_datawrite_start_enabled()) { + char *p = f2fs_kmalloc(F2FS_I_SB(inode), + PATH_MAX, GFP_KERNEL); + char *path; + + if (!p) + goto skip_write_trace; + path = dentry_path_raw(file_dentry(iocb->ki_filp), + p, PATH_MAX); + if (IS_ERR(path)) { + kfree(p); + goto skip_write_trace; + } + trace_f2fs_datawrite_start(inode, orig_pos, orig_count, + current->pid, path, current->comm); + kfree(p); + } +skip_write_trace: /* Do the actual write. */ ret = dio ? f2fs_dio_write_iter(iocb, from, &may_need_sync): f2fs_buffered_write_iter(iocb, from); + if (trace_f2fs_datawrite_end_enabled()) + trace_f2fs_datawrite_end(inode, orig_pos, ret); + } + /* Don't leave any preallocated blocks around past i_size. */ if (preallocated && i_size_read(inode) < target_size) { f2fs_down_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); |