1 What: /sys/fs/f2fs/<disk>/gc_max_sleep_time
3 Contact: "Namjae Jeon" <namjae.jeon@samsung.com>
4 Description: Controls the maximum sleep time for gc_thread. Time
7 What: /sys/fs/f2fs/<disk>/gc_min_sleep_time
9 Contact: "Namjae Jeon" <namjae.jeon@samsung.com>
10 Description: Controls the minimum sleep time for gc_thread. Time
13 What: /sys/fs/f2fs/<disk>/gc_no_gc_sleep_time
15 Contact: "Namjae Jeon" <namjae.jeon@samsung.com>
16 Description: Controls the default sleep time for gc_thread. Time
19 What: /sys/fs/f2fs/<disk>/gc_idle
21 Contact: "Namjae Jeon" <namjae.jeon@samsung.com>
22 Description: Controls the victim selection policy for garbage collection.
23 Setting gc_idle = 0(default) will disable this option. Setting:
25 =========== ===============================================
26 gc_idle = 1 will select the Cost Benefit approach & setting
27 gc_idle = 2 will select the greedy approach & setting
28 gc_idle = 3 will select the age-threshold based approach.
29 =========== ===============================================
31 What: /sys/fs/f2fs/<disk>/reclaim_segments
33 Contact: "Jaegeuk Kim" <jaegeuk.kim@samsung.com>
34 Description: This parameter controls the number of prefree segments to be
35 reclaimed. If the number of prefree segments is larger than
36 the number of segments in the proportion to the percentage
37 over total volume size, f2fs tries to conduct checkpoint to
38 reclaim the prefree segments to free segments.
39 By default, 5% over total # of segments.
41 What: /sys/fs/f2fs/<disk>/main_blkaddr
43 Contact: "Ramon Pantin" <pantin@google.com>
44 Description: Shows first block address of MAIN area.
46 What: /sys/fs/f2fs/<disk>/ipu_policy
48 Contact: "Jaegeuk Kim" <jaegeuk.kim@samsung.com>
49 Description: Controls the in-place-update policy.
50 updates in f2fs. User can set:
52 ==== =================
56 0x08 F2FS_IPU_SSR_UTIL
60 0x80 F2FS_IPU_HONOR_OPU_WRITE
61 ==== =================
63 Refer segment.h for details.
65 What: /sys/fs/f2fs/<disk>/min_ipu_util
67 Contact: "Jaegeuk Kim" <jaegeuk.kim@samsung.com>
68 Description: Controls the FS utilization condition for the in-place-update
69 policies. It is used by F2FS_IPU_UTIL and F2FS_IPU_SSR_UTIL policies.
71 What: /sys/fs/f2fs/<disk>/min_fsync_blocks
73 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
74 Description: Controls the dirty page count condition for the in-place-update
77 What: /sys/fs/f2fs/<disk>/min_seq_blocks
79 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
80 Description: Controls the dirty page count condition for batched sequential
83 What: /sys/fs/f2fs/<disk>/min_hot_blocks
85 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
86 Description: Controls the dirty page count condition for redefining hot data.
88 What: /sys/fs/f2fs/<disk>/min_ssr_sections
90 Contact: "Chao Yu" <yuchao0@huawei.com>
91 Description: Controls the free section threshold to trigger SSR allocation.
92 If this is large, SSR mode will be enabled early.
94 What: /sys/fs/f2fs/<disk>/max_small_discards
96 Contact: "Jaegeuk Kim" <jaegeuk.kim@samsung.com>
97 Description: Controls the issue rate of discard commands that consist of small
98 blocks less than 2MB. The candidates to be discarded are cached until
99 checkpoint is triggered, and issued during the checkpoint.
100 By default, it is disabled with 0.
102 What: /sys/fs/f2fs/<disk>/max_discard_request
104 Contact: "Konstantin Vyshetsky" <vkon@google.com>
105 Description: Controls the number of discards a thread will issue at a time.
106 Higher number will allow the discard thread to finish its work
107 faster, at the cost of higher latency for incomming I/O.
109 What: /sys/fs/f2fs/<disk>/min_discard_issue_time
111 Contact: "Konstantin Vyshetsky" <vkon@google.com>
112 Description: Controls the interval the discard thread will wait between
113 issuing discard requests when there are discards to be issued and
114 no I/O aware interruptions occur.
116 What: /sys/fs/f2fs/<disk>/mid_discard_issue_time
118 Contact: "Konstantin Vyshetsky" <vkon@google.com>
119 Description: Controls the interval the discard thread will wait between
120 issuing discard requests when there are discards to be issued and
121 an I/O aware interruption occurs.
123 What: /sys/fs/f2fs/<disk>/max_discard_issue_time
125 Contact: "Konstantin Vyshetsky" <vkon@google.com>
126 Description: Controls the interval the discard thread will wait when there are
127 no discard operations to be issued.
129 What: /sys/fs/f2fs/<disk>/discard_granularity
131 Contact: "Chao Yu" <yuchao0@huawei.com>
132 Description: Controls discard granularity of inner discard thread. Inner thread
133 will not issue discards with size that is smaller than granularity.
134 The unit size is one block(4KB), now only support configuring
135 in range of [1, 512]. Default value is 4(=16KB).
137 What: /sys/fs/f2fs/<disk>/umount_discard_timeout
139 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
140 Description: Set timeout to issue discard commands during umount.
143 What: /sys/fs/f2fs/<disk>/pending_discard
145 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
146 Description: Shows the number of pending discard commands in the queue.
148 What: /sys/fs/f2fs/<disk>/max_victim_search
150 Contact: "Jaegeuk Kim" <jaegeuk.kim@samsung.com>
151 Description: Controls the number of trials to find a victim segment
152 when conducting SSR and cleaning operations. The default value
153 is 4096 which covers 8GB block address range.
155 What: /sys/fs/f2fs/<disk>/migration_granularity
157 Contact: "Chao Yu" <yuchao0@huawei.com>
158 Description: Controls migration granularity of garbage collection on large
159 section, it can let GC move partial segment{s} of one section
160 in one GC cycle, so that dispersing heavy overhead GC to
161 multiple lightweight one.
163 What: /sys/fs/f2fs/<disk>/dir_level
165 Contact: "Jaegeuk Kim" <jaegeuk.kim@samsung.com>
166 Description: Controls the directory level for large directory. If a
167 directory has a number of files, it can reduce the file lookup
168 latency by increasing this dir_level value. Otherwise, it
169 needs to decrease this value to reduce the space overhead.
170 The default value is 0.
172 What: /sys/fs/f2fs/<disk>/ram_thresh
174 Contact: "Jaegeuk Kim" <jaegeuk.kim@samsung.com>
175 Description: Controls the memory footprint used by free nids and cached
176 nat entries. By default, 1 is set, which indicates
179 What: /sys/fs/f2fs/<disk>/batched_trim_sections
181 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
182 Description: Controls the trimming rate in batch mode.
185 What: /sys/fs/f2fs/<disk>/cp_interval
187 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
188 Description: Controls the checkpoint timing, set to 60 seconds by default.
190 What: /sys/fs/f2fs/<disk>/idle_interval
192 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
193 Description: Controls the idle timing of system, if there is no FS operation
194 during given interval.
195 Set to 5 seconds by default.
197 What: /sys/fs/f2fs/<disk>/discard_idle_interval
199 Contact: "Chao Yu" <yuchao0@huawei.com>
200 Contact: "Sahitya Tummala" <stummala@codeaurora.org>
201 Description: Controls the idle timing of discard thread given
205 What: /sys/fs/f2fs/<disk>/gc_idle_interval
207 Contact: "Chao Yu" <yuchao0@huawei.com>
208 Contact: "Sahitya Tummala" <stummala@codeaurora.org>
209 Description: Controls the idle timing for gc path. Set to 5 seconds by default.
211 What: /sys/fs/f2fs/<disk>/iostat_enable
213 Contact: "Chao Yu" <yuchao0@huawei.com>
214 Description: Controls to enable/disable IO stat.
216 What: /sys/fs/f2fs/<disk>/ra_nid_pages
218 Contact: "Chao Yu" <chao2.yu@samsung.com>
219 Description: Controls the count of nid pages to be readaheaded.
220 When building free nids, F2FS reads NAT blocks ahead for
221 speed up. Default is 0.
223 What: /sys/fs/f2fs/<disk>/dirty_nats_ratio
225 Contact: "Chao Yu" <chao2.yu@samsung.com>
226 Description: Controls dirty nat entries ratio threshold, if current
227 ratio exceeds configured threshold, checkpoint will
228 be triggered for flushing dirty nat entries.
230 What: /sys/fs/f2fs/<disk>/lifetime_write_kbytes
232 Contact: "Shuoran Liu" <liushuoran@huawei.com>
233 Description: Shows total written kbytes issued to disk.
235 What: /sys/fs/f2fs/<disk>/features
237 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
238 Description: <deprecated: should use /sys/fs/f2fs/<disk>/feature_list/
239 Shows all enabled features in current device.
241 encryption, blkzoned, extra_attr, projquota, inode_checksum,
242 flexible_inline_xattr, quota_ino, inode_crtime, lost_found,
243 verity, sb_checksum, casefold, readonly, compression, pin_file.
245 What: /sys/fs/f2fs/<disk>/feature_list/
247 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
248 Description: Expand /sys/fs/f2fs/<disk>/features to meet sysfs rule.
249 Supported on-disk features:
250 encryption, block_zoned (aka blkzoned), extra_attr,
251 project_quota (aka projquota), inode_checksum,
252 flexible_inline_xattr, quota_ino, inode_crtime, lost_found,
253 verity, sb_checksum, casefold, readonly, compression.
254 Note that, pin_file is moved into /sys/fs/f2fs/features/.
256 What: /sys/fs/f2fs/features/
258 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
259 Description: Shows all enabled kernel features.
261 encryption, block_zoned, extra_attr, project_quota,
262 inode_checksum, flexible_inline_xattr, quota_ino,
263 inode_crtime, lost_found, verity, sb_checksum,
264 casefold, readonly, compression, test_dummy_encryption_v2,
265 atomic_write, pin_file, encrypted_casefold.
267 What: /sys/fs/f2fs/<disk>/inject_rate
269 Contact: "Sheng Yong" <shengyong1@huawei.com>
270 Description: Controls the injection rate of arbitrary faults.
272 What: /sys/fs/f2fs/<disk>/inject_type
274 Contact: "Sheng Yong" <shengyong1@huawei.com>
275 Description: Controls the injection type of arbitrary faults.
277 What: /sys/fs/f2fs/<disk>/dirty_segments
279 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
280 Description: Shows the number of dirty segments.
282 What: /sys/fs/f2fs/<disk>/reserved_blocks
284 Contact: "Chao Yu" <yuchao0@huawei.com>
285 Description: Controls target reserved blocks in system, the threshold
286 is soft, it could exceed current available user space.
288 What: /sys/fs/f2fs/<disk>/current_reserved_blocks
290 Contact: "Yunlong Song" <yunlong.song@huawei.com>
291 Contact: "Chao Yu" <yuchao0@huawei.com>
292 Description: Shows current reserved blocks in system, it may be temporarily
293 smaller than target_reserved_blocks, but will gradually
294 increase to target_reserved_blocks when more free blocks are
297 What: /sys/fs/f2fs/<disk>/gc_urgent
299 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
300 Description: Do background GC aggressively when set. Set to 0 by default.
301 gc urgent high(1): does GC forcibly in a period of given
302 gc_urgent_sleep_time and ignores I/O idling check. uses greedy
303 GC approach and turns SSR mode on.
304 gc urgent low(2): lowers the bar of checking I/O idling in
305 order to process outstanding discard commands and GC a
306 little bit aggressively. uses cost benefit GC approach.
307 gc urgent mid(3): does GC forcibly in a period of given
308 gc_urgent_sleep_time and executes a mid level of I/O idling check.
309 uses cost benefit GC approach.
311 What: /sys/fs/f2fs/<disk>/gc_urgent_sleep_time
313 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
314 Description: Controls sleep time of GC urgent mode. Set to 500ms by default.
316 What: /sys/fs/f2fs/<disk>/readdir_ra
318 Contact: "Sheng Yong" <shengyong1@huawei.com>
319 Description: Controls readahead inode block in readdir. Enabled by default.
321 What: /sys/fs/f2fs/<disk>/gc_pin_file_thresh
323 Contact: Jaegeuk Kim <jaegeuk@kernel.org>
324 Description: This indicates how many GC can be failed for the pinned
325 file. If it exceeds this, F2FS doesn't guarantee its pinning
326 state. 2048 trials is set by default.
328 What: /sys/fs/f2fs/<disk>/extension_list
330 Contact: "Chao Yu" <yuchao0@huawei.com>
331 Description: Used to control configure extension list:
332 - Query: cat /sys/fs/f2fs/<disk>/extension_list
333 - Add: echo '[h/c]extension' > /sys/fs/f2fs/<disk>/extension_list
334 - Del: echo '[h/c]!extension' > /sys/fs/f2fs/<disk>/extension_list
335 - [h] means add/del hot file extension
336 - [c] means add/del cold file extension
338 What: /sys/fs/f2fs/<disk>/unusable
340 Contact: "Daniel Rosenberg" <drosen@google.com>
341 Description: If checkpoint=disable, it displays the number of blocks that
343 If checkpoint=enable it displays the number of blocks that
344 would be unusable if checkpoint=disable were to be set.
346 What: /sys/fs/f2fs/<disk>/encoding
348 Contact: "Daniel Rosenberg" <drosen@google.com>
349 Description: Displays name and version of the encoding set for the filesystem.
350 If no encoding is set, displays (none)
352 What: /sys/fs/f2fs/<disk>/free_segments
354 Contact: "Hridya Valsaraju" <hridya@google.com>
355 Description: Number of free segments in disk.
357 What: /sys/fs/f2fs/<disk>/cp_foreground_calls
359 Contact: "Hridya Valsaraju" <hridya@google.com>
360 Description: Number of checkpoint operations performed on demand. Available when
361 CONFIG_F2FS_STAT_FS=y.
363 What: /sys/fs/f2fs/<disk>/cp_background_calls
365 Contact: "Hridya Valsaraju" <hridya@google.com>
366 Description: Number of checkpoint operations performed in the background to
367 free segments. Available when CONFIG_F2FS_STAT_FS=y.
369 What: /sys/fs/f2fs/<disk>/gc_foreground_calls
371 Contact: "Hridya Valsaraju" <hridya@google.com>
372 Description: Number of garbage collection operations performed on demand.
373 Available when CONFIG_F2FS_STAT_FS=y.
375 What: /sys/fs/f2fs/<disk>/gc_background_calls
377 Contact: "Hridya Valsaraju" <hridya@google.com>
378 Description: Number of garbage collection operations triggered in background.
379 Available when CONFIG_F2FS_STAT_FS=y.
381 What: /sys/fs/f2fs/<disk>/moved_blocks_foreground
383 Contact: "Hridya Valsaraju" <hridya@google.com>
384 Description: Number of blocks moved by garbage collection in foreground.
385 Available when CONFIG_F2FS_STAT_FS=y.
387 What: /sys/fs/f2fs/<disk>/moved_blocks_background
389 Contact: "Hridya Valsaraju" <hridya@google.com>
390 Description: Number of blocks moved by garbage collection in background.
391 Available when CONFIG_F2FS_STAT_FS=y.
393 What: /sys/fs/f2fs/<disk>/avg_vblocks
395 Contact: "Hridya Valsaraju" <hridya@google.com>
396 Description: Average number of valid blocks.
397 Available when CONFIG_F2FS_STAT_FS=y.
399 What: /sys/fs/f2fs/<disk>/mounted_time_sec
401 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
402 Description: Show the mounted time in secs of this partition.
404 What: /sys/fs/f2fs/<disk>/data_io_flag
406 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
407 Description: Give a way to attach REQ_META|FUA to data writes
408 given temperature-based bits. Now the bits indicate:
410 +-------------------+-------------------+
411 | REQ_META | REQ_FUA |
412 +------+------+-----+------+------+-----+
413 | 5 | 4 | 3 | 2 | 1 | 0 |
414 +------+------+-----+------+------+-----+
415 | Cold | Warm | Hot | Cold | Warm | Hot |
416 +------+------+-----+------+------+-----+
418 What: /sys/fs/f2fs/<disk>/node_io_flag
420 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
421 Description: Give a way to attach REQ_META|FUA to node writes
422 given temperature-based bits. Now the bits indicate:
424 +-------------------+-------------------+
425 | REQ_META | REQ_FUA |
426 +------+------+-----+------+------+-----+
427 | 5 | 4 | 3 | 2 | 1 | 0 |
428 +------+------+-----+------+------+-----+
429 | Cold | Warm | Hot | Cold | Warm | Hot |
430 +------+------+-----+------+------+-----+
432 What: /sys/fs/f2fs/<disk>/iostat_period_ms
434 Contact: "Daeho Jeong" <daehojeong@google.com>
435 Description: Give a way to change iostat_period time. 3secs by default.
436 The new iostat trace gives stats gap given the period.
437 What: /sys/fs/f2fs/<disk>/max_io_bytes
439 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
440 Description: This gives a control to limit the bio size in f2fs.
441 Default is zero, which will follow underlying block layer limit,
442 whereas, if it has a certain bytes value, f2fs won't submit a
443 bio larger than that size.
445 What: /sys/fs/f2fs/<disk>/stat/sb_status
447 Contact: "Chao Yu" <yuchao0@huawei.com>
448 Description: Show status of f2fs superblock in real time.
450 ====== ===================== =================================
451 value sb status macro description
452 0x1 SBI_IS_DIRTY dirty flag for checkpoint
453 0x2 SBI_IS_CLOSE specify unmounting
454 0x4 SBI_NEED_FSCK need fsck.f2fs to fix
455 0x8 SBI_POR_DOING recovery is doing or not
456 0x10 SBI_NEED_SB_WRITE need to recover superblock
457 0x20 SBI_NEED_CP need to checkpoint
458 0x40 SBI_IS_SHUTDOWN shutdown by ioctl
459 0x80 SBI_IS_RECOVERED recovered orphan/data
460 0x100 SBI_CP_DISABLED CP was disabled last mount
461 0x200 SBI_CP_DISABLED_QUICK CP was disabled quickly
462 0x400 SBI_QUOTA_NEED_FLUSH need to flush quota info in CP
463 0x800 SBI_QUOTA_SKIP_FLUSH skip flushing quota in current CP
464 0x1000 SBI_QUOTA_NEED_REPAIR quota file may be corrupted
465 0x2000 SBI_IS_RESIZEFS resizefs is in process
466 0x4000 SBI_IS_FREEZING freefs is in process
467 ====== ===================== =================================
469 What: /sys/fs/f2fs/<disk>/ckpt_thread_ioprio
471 Contact: "Daeho Jeong" <daehojeong@google.com>
472 Description: Give a way to change checkpoint merge daemon's io priority.
473 Its default value is "be,3", which means "BE" I/O class and
474 I/O priority "3". We can select the class between "rt" and "be",
475 and set the I/O priority within valid range of it. "," delimiter
476 is necessary in between I/O class and priority number.
478 What: /sys/fs/f2fs/<disk>/ovp_segments
480 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
481 Description: Shows the number of overprovision segments.
483 What: /sys/fs/f2fs/<disk>/compr_written_block
485 Contact: "Daeho Jeong" <daehojeong@google.com>
486 Description: Show the block count written after compression since mount. Note
487 that when the compressed blocks are deleted, this count doesn't
488 decrease. If you write "0" here, you can initialize
489 compr_written_block and compr_saved_block to "0".
491 What: /sys/fs/f2fs/<disk>/compr_saved_block
493 Contact: "Daeho Jeong" <daehojeong@google.com>
494 Description: Show the saved block count with compression since mount. Note
495 that when the compressed blocks are deleted, this count doesn't
496 decrease. If you write "0" here, you can initialize
497 compr_written_block and compr_saved_block to "0".
499 What: /sys/fs/f2fs/<disk>/compr_new_inode
501 Contact: "Daeho Jeong" <daehojeong@google.com>
502 Description: Show the count of inode newly enabled for compression since mount.
503 Note that when the compression is disabled for the files, this count
504 doesn't decrease. If you write "0" here, you can initialize
505 compr_new_inode to "0".
507 What: /sys/fs/f2fs/<disk>/atgc_candidate_ratio
509 Contact: "Chao Yu" <yuchao0@huawei.com>
510 Description: When ATGC is on, it controls candidate ratio in order to limit total
511 number of potential victim in all candidates, the value should be in
512 range of [0, 100], by default it was initialized as 20(%).
514 What: /sys/fs/f2fs/<disk>/atgc_candidate_count
516 Contact: "Chao Yu" <yuchao0@huawei.com>
517 Description: When ATGC is on, it controls candidate count in order to limit total
518 number of potential victim in all candidates, by default it was
519 initialized as 10 (sections).
521 What: /sys/fs/f2fs/<disk>/atgc_age_weight
523 Contact: "Chao Yu" <yuchao0@huawei.com>
524 Description: When ATGC is on, it controls age weight to balance weight proportion
525 in between aging and valid blocks, the value should be in range of
526 [0, 100], by default it was initialized as 60(%).
528 What: /sys/fs/f2fs/<disk>/atgc_age_threshold
530 Contact: "Chao Yu" <yuchao0@huawei.com>
531 Description: When ATGC is on, it controls age threshold to bypass GCing young
532 candidates whose age is not beyond the threshold, by default it was
533 initialized as 604800 seconds (equals to 7 days).
535 What: /sys/fs/f2fs/<disk>/gc_reclaimed_segments
537 Contact: "Daeho Jeong" <daehojeong@google.com>
538 Description: Show how many segments have been reclaimed by GC during a specific
539 GC mode (0: GC normal, 1: GC idle CB, 2: GC idle greedy,
540 3: GC idle AT, 4: GC urgent high, 5: GC urgent low 6: GC urgent mid)
541 You can re-initialize this value to "0".
543 What: /sys/fs/f2fs/<disk>/gc_segment_mode
545 Contact: "Daeho Jeong" <daehojeong@google.com>
546 Description: You can control for which gc mode the "gc_reclaimed_segments" node shows.
547 Refer to the description of the modes in "gc_reclaimed_segments".
549 What: /sys/fs/f2fs/<disk>/seq_file_ra_mul
551 Contact: "Daeho Jeong" <daehojeong@google.com>
552 Description: You can control the multiplier value of bdi device readahead window size
553 between 2 (default) and 256 for POSIX_FADV_SEQUENTIAL advise option.
555 What: /sys/fs/f2fs/<disk>/max_fragment_chunk
557 Contact: "Daeho Jeong" <daehojeong@google.com>
558 Description: With "mode=fragment:block" mount options, we can scatter block allocation.
559 f2fs will allocate 1..<max_fragment_chunk> blocks in a chunk and make a hole
560 in the length of 1..<max_fragment_hole> by turns. This value can be set
561 between 1..512 and the default value is 4.
563 What: /sys/fs/f2fs/<disk>/max_fragment_hole
565 Contact: "Daeho Jeong" <daehojeong@google.com>
566 Description: With "mode=fragment:block" mount options, we can scatter block allocation.
567 f2fs will allocate 1..<max_fragment_chunk> blocks in a chunk and make a hole
568 in the length of 1..<max_fragment_hole> by turns. This value can be set
569 between 1..512 and the default value is 4.
571 What: /sys/fs/f2fs/<disk>/gc_urgent_high_remaining
573 Contact: "Daeho Jeong" <daehojeong@google.com>
574 Description: You can set the trial count limit for GC urgent high mode with this value.
575 If GC thread gets to the limit, the mode will turn back to GC normal mode.
576 By default, the value is zero, which means there is no limit like before.
578 What: /sys/fs/f2fs/<disk>/max_roll_forward_node_blocks
580 Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
581 Description: Controls max # of node block writes to be used for roll forward
582 recovery. This can limit the roll forward recovery time.