[f2fs-dev] [PATCH AUTOSEL 4.19 05/37] f2fs: fix to do sanity check on valid node/block count
From: Chao Yu
[ Upstream commit 7b63f72f73af57bb040f03b9713ec9979d8911f4 ]
As Jungyeon reported in bugzilla:
https://bugzilla.kernel.org/show_bug.cgi?id=203229
- Overview
When mounting the attached crafted image, following errors are reported.
Additionally, it hangs on sync after trying to mount it.
The image is intentionally fuzzed from a normal f2fs image for testing.
Compile options for F2FS are as follows.
CONFIG_F2FS_FS=y
CONFIG_F2FS_STAT_FS=y
CONFIG_F2FS_FS_XATTR=y
CONFIG_F2FS_FS_POSIX_ACL=y
CONFIG_F2FS_CHECK_FS=y
- Reproduces
mkdir test
mount -t f2fs tmp.img test
sync
- Kernel message
kernel BUG at fs/f2fs/recovery.c:591!
RIP: 0010:recover_data+0x12d8/0x1780
Call Trace:
f2fs_recover_fsync_data+0x613/0x710
f2fs_fill_super+0x1043/0x1aa0
mount_bdev+0x16d/0x1a0
mount_fs+0x4a/0x170
vfs_kern_mount+0x5d/0x100
do_mount+0x200/0xcf0
ksys_mount+0x79/0xc0
__x64_sys_mount+0x1c/0x20
do_syscall_64+0x43/0xf0
entry_SYSCALL_64_after_hwframe+0x44/0xa9
With corrupted image wihch has out-of-range valid node/block count, during
recovery, once we failed due to no free space, it will trigger kernel
panic.
Adding sanity check on valid node/block count in f2fs_sanity_check_ckpt()
to detect such condition, so that potential panic can be avoided.
Signed-off-by: Chao Yu
Signed-off-by: Jaegeuk Kim
Signed-off-by: Sasha Levin
---
fs/f2fs/super.c | 21 -
1 file changed, 20 insertions(+), 1 deletion(-)
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 6851afc3bf805..3fc96cffe6ac5 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -2344,7 +2344,8 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi)
unsigned int log_blocks_per_seg;
unsigned int segment_count_main;
unsigned int cp_pack_start_sum, cp_payload;
- block_t user_block_count;
+ block_t user_block_count, valid_user_blocks;
+ block_t avail_node_count, valid_node_count;
int i, j;
total = le32_to_cpu(raw_super->segment_count);
@@ -2379,6 +2380,24 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi)
return 1;
}
+ valid_user_blocks = le64_to_cpu(ckpt->valid_block_count);
+ if (valid_user_blocks > user_block_count) {
+ f2fs_msg(sbi->sb, KERN_ERR,
+ "Wrong valid_user_blocks: %u, user_block_count: %u",
+ valid_user_blocks, user_block_count);
+ return 1;
+ }
+
+ valid_node_count = le32_to_cpu(ckpt->valid_node_count);
+ avail_node_count = sbi->total_node_count - sbi->nquota_files -
+ F2FS_RESERVED_NODE_NUM;
+ if (valid_node_count > avail_node_count) {
+ f2fs_msg(sbi->sb, KERN_ERR,
+ "Wrong valid_node_count: %u, avail_node_count: %u",
+ valid_node_count, avail_node_count);
+ return 1;
+ }
+
main_segs = le32_to_cpu(raw_super->segment_count_main);
blocks_per_seg = sbi->blocks_per_seg;
--
2.20.1
___
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Re: [f2fs-dev] [bug report] compiler warning: fs/f2fs/node.c: In function ‘__set_nat_cache_dirty’: ‘head’ may be used uninitialized
On 2019/10/24 17:54, Ardelean, Alexandru wrote: > On Thu, 2019-10-24 at 17:12 +0800, Chao Yu wrote: >> [External] >> >> On 2019/10/23 22:02, Ardelean, Alexandru wrote: >>> Seems to have been introduced via: >>> >>> >>> >>> commit 780de47cf6cb5f524cd98ec8ffbffc3da5696e17 >>> Author: Chao Yu >>> Date: Tue Mar 20 23:08:30 2018 +0800 >>> >>> f2fs: don't track new nat entry in nat set >>> >>> Nat entry set is used only in checkpoint(), and during checkpoint() >>> we >>> won't flush new nat entry with unallocated address, so we don't >>> need to >>> add new nat entry into nat set, then nat_entry_set::entry_cnt can >>> indicate actual entry count we need to flush in checkpoint(). >>> >>> Signed-off-by: Yunlei He >>> Signed-off-by: Chao Yu >>> Signed-off-by: Jaegeuk Kim >>> >>> >>> Compiler warning is: >>> >>> >>> CC fs/f2fs/node.o >>> In file included from ./include/linux/wait.h:7:0, >>> from ./include/linux/wait_bit.h:8, >>> from ./include/linux/fs.h:6, >>> from fs/f2fs/node.c:11: >>> fs/f2fs/node.c: In function ‘__set_nat_cache_dirty’: >>> ./include/linux/list.h:63:13: error: ‘head’ may be used uninitialized >>> in >>> this function [-Werror=maybe-uninitialized] >>> next->prev = new; >>> ^ >>> fs/f2fs/node.c:238:24: note: ‘head’ was declared here >>> struct nat_entry_set *head; >> >> That's not correct, @head will only be assigned and used if new_ne equals >> NULL. > > Ack. > I admit that I don't understand the code, and don't claim to understand it. > > This may be just a weird compiler issue. > I thought I'd send it just as a heads-up. I think that's the right thing to do. > I saw this on a Raspberry Pi branch [after we enabled warnings as errors]: > https://travis-ci.org/analogdevicesinc/linux/jobs/601844926#L1208 > > Looking in the latest f2fs/dev[-test] tree, it looks like the code is > similar as in 4.19. > https://github.com/analogdevicesinc/linux/blob/rpi-4.19.y/fs/f2fs/node.c#L235 > > Could be that the RPi branch has some more compiler-stuff enabled. > > In any case, feel free to disregard this. > We will see how we fix this on our end for that branch specifically. That would make sense, let me know if you have any other suspicious compiler warnings. :) Thanks, > > Thanks > Alex > >> >> Thanks, >> >>> ^ >>> cc1: all warnings being treated as errors >>> >>> >>> Thanks >>> Alex >>> >>> ___ >>> Linux-f2fs-devel mailing list >>> Linux-f2fs-devel@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel >>> ___ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Re: [f2fs-dev] [PATCH 2/2] f2fs: Add f2fs stats to sysfs
On 2019/10/25 11:51, Hridya Valsaraju wrote: > On Thu, Oct 24, 2019 at 2:26 AM Chao Yu wrote: >> >> On 2019/10/24 5:48, Hridya Valsaraju wrote: >>> Currently f2fs stats are only available from /d/f2fs/status. This patch >>> adds some of the f2fs stats to sysfs so that they are accessible even >>> when debugfs is not mounted. >> >> Why don't we mount debugfs first? > > Thank you for taking a look at the patch Chao. We will not be mounting > debugfs for security reasons. Hi, Hridya, May I ask is there any use case for those new entries? So many sysfs entries exist, if there is real use case, how about backuping entire /d/f2fs/status entry into /proc/fs/f2fs// directory rather than adding some of stats as a single entry in sysfs directory? Thanks, > > Regards, > Hridya > >> >> Thanks, > . > ___ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Re: [f2fs-dev] [PATCH 1/2] f2fs: support aligned pinned file
On 10/24, Chao Yu wrote:
> Hi Jaegeuk,
>
> On 2019/10/23 1:16, Jaegeuk Kim wrote:
> > This patch supports 2MB-aligned pinned file, which can guarantee no GC at
> > all
> > by allocating fully valid 2MB segment.
> >
> > Signed-off-by: Jaegeuk Kim
> > ---
> > fs/f2fs/f2fs.h | 4 +++-
> > fs/f2fs/file.c | 39 ++-
> > fs/f2fs/recovery.c | 2 +-
> > fs/f2fs/segment.c | 21 -
> > fs/f2fs/segment.h | 2 ++
> > fs/f2fs/super.c| 1 +
> > fs/f2fs/sysfs.c| 2 ++
> > 7 files changed, 63 insertions(+), 8 deletions(-)
> >
> > diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
> > index ca342f4c7db1..c681f51e351b 100644
> > --- a/fs/f2fs/f2fs.h
> > +++ b/fs/f2fs/f2fs.h
> > @@ -890,6 +890,7 @@ enum {
> > CURSEG_WARM_NODE, /* direct node blocks of normal files */
> > CURSEG_COLD_NODE, /* indirect node blocks */
> > NO_CHECK_TYPE,
> > + CURSEG_COLD_DATA_PINNED,/* cold data for pinned file */
> > };
> >
> > struct flush_cmd {
> > @@ -1301,6 +1302,7 @@ struct f2fs_sb_info {
> >
> > /* threshold for gc trials on pinned files */
> > u64 gc_pin_file_threshold;
> > + struct rw_semaphore pin_sem;
> >
> > /* maximum # of trials to find a victim segment for SSR and GC */
> > unsigned int max_victim_search;
> > @@ -3116,7 +3118,7 @@ void f2fs_release_discard_addrs(struct f2fs_sb_info
> > *sbi);
> > int f2fs_npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra);
> > void allocate_segment_for_resize(struct f2fs_sb_info *sbi, int type,
> > unsigned int start, unsigned int end);
> > -void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi);
> > +void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi, int type);
> > int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range);
> > bool f2fs_exist_trim_candidates(struct f2fs_sb_info *sbi,
> > struct cp_control *cpc);
> > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
> > index 29bc0a542759..f6c038e8a6a7 100644
> > --- a/fs/f2fs/file.c
> > +++ b/fs/f2fs/file.c
> > @@ -1545,12 +1545,41 @@ static int expand_inode_data(struct inode *inode,
> > loff_t offset,
> > if (off_end)
> > map.m_len++;
> >
> > - if (f2fs_is_pinned_file(inode))
> > - map.m_seg_type = CURSEG_COLD_DATA;
> > + if (!map.m_len)
> > + return 0;
> > +
> > + if (f2fs_is_pinned_file(inode)) {
> > + block_t len = (map.m_len >> sbi->log_blocks_per_seg) <<
> > + sbi->log_blocks_per_seg;
> > + block_t done = 0;
> > +
> > + if (map.m_len % sbi->blocks_per_seg)
> > + len += sbi->blocks_per_seg;
> >
> > - err = f2fs_map_blocks(inode, , 1, (f2fs_is_pinned_file(inode) ?
> > - F2FS_GET_BLOCK_PRE_DIO :
> > - F2FS_GET_BLOCK_PRE_AIO));
> > + map.m_len = sbi->blocks_per_seg;
> > +next_alloc:
> > + mutex_lock(>gc_mutex);
> > + err = f2fs_gc(sbi, true, false, NULL_SEGNO);
> > + if (err && err != -ENODATA && err != -EAGAIN)
> > + goto out_err;
>
> To grab enough free space?
>
> Shouldn't we call
>
> if (has_not_enough_free_secs(sbi, 0, 0)) {
> mutex_lock(>gc_mutex);
> f2fs_gc(sbi, false, false, NULL_SEGNO);
> }
The above calls gc all the time. Do we need this?
>
> > +
> > + down_write(>pin_sem);
> > + map.m_seg_type = CURSEG_COLD_DATA_PINNED;
> > + f2fs_allocate_new_segments(sbi, CURSEG_COLD_DATA);
> > + err = f2fs_map_blocks(inode, , 1, F2FS_GET_BLOCK_PRE_DIO);
> > + up_write(>pin_sem);
> > +
> > + done += map.m_len;
> > + len -= map.m_len;
> > + map.m_lblk += map.m_len;
> > + if (!err && len)
> > + goto next_alloc;
> > +
> > + map.m_len = done;
> > + } else {
> > + err = f2fs_map_blocks(inode, , 1, F2FS_GET_BLOCK_PRE_AIO);
> > + }
> > +out_err:
> > if (err) {
> > pgoff_t last_off;
> >
> > diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
> > index 783773e4560d..76477f71d4ee 100644
> > --- a/fs/f2fs/recovery.c
> > +++ b/fs/f2fs/recovery.c
> > @@ -711,7 +711,7 @@ static int recover_data(struct f2fs_sb_info *sbi,
> > struct list_head *inode_list,
> > f2fs_put_page(page, 1);
> > }
> > if (!err)
> > - f2fs_allocate_new_segments(sbi);
> > + f2fs_allocate_new_segments(sbi, NO_CHECK_TYPE);
> > return err;
> > }
> >
> > diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
> > index 25c750cd0272..253d72c2663c 100644
> > --- a/fs/f2fs/segment.c
> > +++ b/fs/f2fs/segment.c
> > @@ -2690,7 +2690,7 @@ void allocate_segment_for_resize(struct f2fs_sb_info
> > *sbi, int type,
> > up_read(_I(sbi)->curseg_lock);
> > }
> >
>
Re: [f2fs-dev] [PATCH 2/2] f2fs: Add f2fs stats to sysfs
On 10/25, Chao Yu wrote: > On 2019/10/25 11:51, Hridya Valsaraju wrote: > > On Thu, Oct 24, 2019 at 2:26 AM Chao Yu wrote: > >> > >> On 2019/10/24 5:48, Hridya Valsaraju wrote: > >>> Currently f2fs stats are only available from /d/f2fs/status. This patch > >>> adds some of the f2fs stats to sysfs so that they are accessible even > >>> when debugfs is not mounted. > >> > >> Why don't we mount debugfs first? > > > > Thank you for taking a look at the patch Chao. We will not be mounting > > debugfs for security reasons. > > Hi, Hridya, > > May I ask is there any use case for those new entries? > > So many sysfs entries exist, if there is real use case, how about backuping > entire /d/f2fs/status entry into /proc/fs/f2fs// directory rather than > adding some of stats as a single entry in sysfs directory? These will be useful to keep a track on f2fs health status by one value per entry, which doesn't require user-land parsing stuff. Of course, Android can exploit them by IdleMaint, rollback feature, and so on. > > Thanks, > > > > > Regards, > > Hridya > > > >> > >> Thanks, > > . > > ___ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Re: [f2fs-dev] [PATCH 1/2] f2fs: delete duplicate information on sysfs nodes
On 10/24, Chao Yu wrote: > On 2019/10/24 5:48, Hridya Valsaraju wrote: > > This patch merges the sysfs node documentation present in > > Documentation/filesystems/f2fs.txt and > > Documentation/ABI/testing/sysfs-fs-f2fs > > and deletes the duplicate information from > > Documentation/filesystems/f2fs.txt. This is to prevent having to update > > both files when a new sysfs node is added for f2fs. > > The patch also makes minor formatting changes to > > Documentation/ABI/testing/sysfs-fs-f2fs. > > Jaegeuk, any particular reason to add duplicated description on f2fs.txt > previously? Not at all, thus, I asked Hridya to consolidate them. :P > > Thanks, ___ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Re: [f2fs-dev] [PATCH 1/2] f2fs: delete duplicate information on sysfs nodes
On 2019/10/26 2:23, Jaegeuk Kim wrote: > On 10/24, Chao Yu wrote: >> On 2019/10/24 5:48, Hridya Valsaraju wrote: >>> This patch merges the sysfs node documentation present in >>> Documentation/filesystems/f2fs.txt and >>> Documentation/ABI/testing/sysfs-fs-f2fs >>> and deletes the duplicate information from >>> Documentation/filesystems/f2fs.txt. This is to prevent having to update >>> both files when a new sysfs node is added for f2fs. >>> The patch also makes minor formatting changes to >>> Documentation/ABI/testing/sysfs-fs-f2fs. >> >> Jaegeuk, any particular reason to add duplicated description on f2fs.txt >> previously? > > Not at all, thus, I asked Hridya to consolidate them. :P Alright... Reviewed-by: Chao Yu Thanks, > >> >> Thanks, > . > ___ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Re: [f2fs-dev] [PATCH 2/2] f2fs: Add f2fs stats to sysfs
On 2019/10/26 2:22, Jaegeuk Kim wrote: > On 10/25, Chao Yu wrote: >> On 2019/10/25 11:51, Hridya Valsaraju wrote: >>> On Thu, Oct 24, 2019 at 2:26 AM Chao Yu wrote: On 2019/10/24 5:48, Hridya Valsaraju wrote: > Currently f2fs stats are only available from /d/f2fs/status. This patch > adds some of the f2fs stats to sysfs so that they are accessible even > when debugfs is not mounted. Why don't we mount debugfs first? >>> >>> Thank you for taking a look at the patch Chao. We will not be mounting >>> debugfs for security reasons. >> >> Hi, Hridya, >> >> May I ask is there any use case for those new entries? >> >> So many sysfs entries exist, if there is real use case, how about backuping >> entire /d/f2fs/status entry into /proc/fs/f2fs// directory rather than >> adding some of stats as a single entry in sysfs directory? > > These will be useful to keep a track on f2fs health status by one value > per entry, which doesn't require user-land parsing stuff. Of course, Android > can exploit them by IdleMaint, rollback feature, and so on. Alright, I suggest to add a sub-directory for those statistic entries, we can manage them more easily isolated from those existed switch entries. Thanks, > >> >> Thanks, >> >>> >>> Regards, >>> Hridya >>> Thanks, >>> . >>> > . > ___ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Re: [f2fs-dev] [PATCH 1/2] f2fs: support aligned pinned file
On 2019/10/26 2:18, Jaegeuk Kim wrote:
> On 10/24, Chao Yu wrote:
>> Hi Jaegeuk,
>>
>> On 2019/10/23 1:16, Jaegeuk Kim wrote:
>>> This patch supports 2MB-aligned pinned file, which can guarantee no GC at
>>> all
>>> by allocating fully valid 2MB segment.
>>>
>>> Signed-off-by: Jaegeuk Kim
>>> ---
>>> fs/f2fs/f2fs.h | 4 +++-
>>> fs/f2fs/file.c | 39 ++-
>>> fs/f2fs/recovery.c | 2 +-
>>> fs/f2fs/segment.c | 21 -
>>> fs/f2fs/segment.h | 2 ++
>>> fs/f2fs/super.c| 1 +
>>> fs/f2fs/sysfs.c| 2 ++
>>> 7 files changed, 63 insertions(+), 8 deletions(-)
>>>
>>> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
>>> index ca342f4c7db1..c681f51e351b 100644
>>> --- a/fs/f2fs/f2fs.h
>>> +++ b/fs/f2fs/f2fs.h
>>> @@ -890,6 +890,7 @@ enum {
>>> CURSEG_WARM_NODE, /* direct node blocks of normal files */
>>> CURSEG_COLD_NODE, /* indirect node blocks */
>>> NO_CHECK_TYPE,
>>> + CURSEG_COLD_DATA_PINNED,/* cold data for pinned file */
>>> };
>>>
>>> struct flush_cmd {
>>> @@ -1301,6 +1302,7 @@ struct f2fs_sb_info {
>>>
>>> /* threshold for gc trials on pinned files */
>>> u64 gc_pin_file_threshold;
>>> + struct rw_semaphore pin_sem;
>>>
>>> /* maximum # of trials to find a victim segment for SSR and GC */
>>> unsigned int max_victim_search;
>>> @@ -3116,7 +3118,7 @@ void f2fs_release_discard_addrs(struct f2fs_sb_info
>>> *sbi);
>>> int f2fs_npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra);
>>> void allocate_segment_for_resize(struct f2fs_sb_info *sbi, int type,
>>> unsigned int start, unsigned int end);
>>> -void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi);
>>> +void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi, int type);
>>> int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range);
>>> bool f2fs_exist_trim_candidates(struct f2fs_sb_info *sbi,
>>> struct cp_control *cpc);
>>> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
>>> index 29bc0a542759..f6c038e8a6a7 100644
>>> --- a/fs/f2fs/file.c
>>> +++ b/fs/f2fs/file.c
>>> @@ -1545,12 +1545,41 @@ static int expand_inode_data(struct inode *inode,
>>> loff_t offset,
>>> if (off_end)
>>> map.m_len++;
>>>
>>> - if (f2fs_is_pinned_file(inode))
>>> - map.m_seg_type = CURSEG_COLD_DATA;
>>> + if (!map.m_len)
>>> + return 0;
>>> +
>>> + if (f2fs_is_pinned_file(inode)) {
>>> + block_t len = (map.m_len >> sbi->log_blocks_per_seg) <<
>>> + sbi->log_blocks_per_seg;
>>> + block_t done = 0;
>>> +
>>> + if (map.m_len % sbi->blocks_per_seg)
>>> + len += sbi->blocks_per_seg;
>>>
>>> - err = f2fs_map_blocks(inode, , 1, (f2fs_is_pinned_file(inode) ?
>>> - F2FS_GET_BLOCK_PRE_DIO :
>>> - F2FS_GET_BLOCK_PRE_AIO));
>>> + map.m_len = sbi->blocks_per_seg;
>>> +next_alloc:
>>> + mutex_lock(>gc_mutex);
>>> + err = f2fs_gc(sbi, true, false, NULL_SEGNO);
>>> + if (err && err != -ENODATA && err != -EAGAIN)
>>> + goto out_err;
>>
>> To grab enough free space?
>>
>> Shouldn't we call
>>
>> if (has_not_enough_free_secs(sbi, 0, 0)) {
>> mutex_lock(>gc_mutex);
>> f2fs_gc(sbi, false, false, NULL_SEGNO);
>> }
>
> The above calls gc all the time. Do we need this?
Hmmm... my concern is why we need to run foreground GC even if there is enough
free space..
>
>>
>>> +
>>> + down_write(>pin_sem);
>>> + map.m_seg_type = CURSEG_COLD_DATA_PINNED;
>>> + f2fs_allocate_new_segments(sbi, CURSEG_COLD_DATA);
>>> + err = f2fs_map_blocks(inode, , 1, F2FS_GET_BLOCK_PRE_DIO);
>>> + up_write(>pin_sem);
>>> +
>>> + done += map.m_len;
>>> + len -= map.m_len;
>>> + map.m_lblk += map.m_len;
>>> + if (!err && len)
>>> + goto next_alloc;
>>> +
>>> + map.m_len = done;
>>> + } else {
>>> + err = f2fs_map_blocks(inode, , 1, F2FS_GET_BLOCK_PRE_AIO);
>>> + }
>>> +out_err:
>>> if (err) {
>>> pgoff_t last_off;
>>>
>>> diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
>>> index 783773e4560d..76477f71d4ee 100644
>>> --- a/fs/f2fs/recovery.c
>>> +++ b/fs/f2fs/recovery.c
>>> @@ -711,7 +711,7 @@ static int recover_data(struct f2fs_sb_info *sbi,
>>> struct list_head *inode_list,
>>> f2fs_put_page(page, 1);
>>> }
>>> if (!err)
>>> - f2fs_allocate_new_segments(sbi);
>>> + f2fs_allocate_new_segments(sbi, NO_CHECK_TYPE);
>>> return err;
>>> }
>>>
>>> diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
>>> index 25c750cd0272..253d72c2663c 100644
>>> --- a/fs/f2fs/segment.c
>>> +++ b/fs/f2fs/segment.c
>>> @@
