Previously, when f2fs finds which temp bio cache owns the target page,
it will flush all the three temp bio caches, but we only need to flush
one single bio cache indeed, which can help to keep bio merged.
Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In get_more_blocks(), we may override @create as below code:
create = dio->op == REQ_OP_WRITE;
if (dio->flags & DIO_SKIP_HOLES) {
if (fs_startblk <= ((i_size_read(dio->inode) - 1) >>
i_blkbits))
create = 0;
}
But in f2fs_map_blocks(), we only trigger f2fs_balance_fs() if @create
is 1, so in LFS mode, dio overwrite under LFS mode can easily run out
of free segments, result in below panic.
Call Trace:
allocate_segment_by_default+0xa8/0x270 [f2fs]
f2fs_allocate_data_block+0x1ea/0x5c0 [f2fs]
__allocate_data_block+0x306/0x480 [f2fs]
f2fs_map_blocks+0x6f6/0x920 [f2fs]
__get_data_block+0x4f/0xb0 [f2fs]
get_data_block_dio_write+0x50/0x60 [f2fs]
do_blockdev_direct_IO+0xcd5/0x21e0
__blockdev_direct_IO+0x3a/0x3c
f2fs_direct_IO+0x1ff/0x4a0 [f2fs]
generic_file_direct_write+0xd9/0x160
__generic_file_write_iter+0xbb/0x1e0
f2fs_file_write_iter+0xaf/0x220 [f2fs]
__vfs_write+0xd0/0x130
vfs_write+0xb2/0x1b0
SyS_pwrite64+0x69/0xa0
? vtime_user_exit+0x29/0x70
do_syscall_64+0x6e/0x160
entry_SYSCALL64_slow_path+0x25/0x25
RIP: new_curseg+0x36f/0x380 [f2fs] RSP: ffffac570393f7a8
So this patch introduces a parameter map.m_may_create to indicate that
f2fs_map_blocks() is called from write or read path, which can give the
right hint to let f2fs_map_blocks() trigger OPU allocation and call
f2fs_balanc_fs() correctly.
BTW, it disables physical address preallocation for direct IO in
f2fs_preallocate_blocks, which is redundant to OPU allocation of
f2fs_map_blocks.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch adds f2fs_dio_submit_bio() to hook submit_io/end_io functions
in direct IO path, in order to account DIO.
Later, we will add this count into is_idle() to let background GC/Discard
thread be aware of DIO.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch move dir data flush to write checkpoint process, by
doing this, it may reduce some time for dir fsync.
pre:
-f2fs_do_sync_file enter
-file_write_and_wait_range <- flush & wait
-write_checkpoint
-do_checkpoint <- wait all
-f2fs_do_sync_file exit
now:
-f2fs_do_sync_file enter
-write_checkpoint
-block_operations <- flush dir & no wait
-do_checkpoint <- wait all
-f2fs_do_sync_file exit
Signed-off-by: Yunlei He <heyunlei@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
f2fs_ioc_gc_range skips blocks_per_seg each time, however, f2fs_gc moves
blocks of section each time, so fix it from segment to section.
Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Add one sysfs entry to control migration granularity of GC in large
section f2fs, it can be tuned to mitigate heavy overhead of migrating
huge number of blocks in large section.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Section is minimal garbage collection unit of f2fs, in zoned block
device, or ancient block mapping flash device, in order to improve
GC efficiency, we can align GC unit to lower device erase unit,
normally, it consists of multiple of segments.
Once background or foreground GC triggers, it brings a large number
of IOs, which will impact user IO, and also occupy cpu/memory resource
intensively.
So, to reduce impact of GC on large size section, this patch supports
subsectional GC, in one cycle of GC, it only migrate partial segment{s}
in victim section. Currently, by default, we use sbi->segs_per_sec as
migration granularity.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Introduce a wrapper __is_large_section() to clean up codes.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In F2FS_HAS_FEATURE(), we will use F2FS_SB(sb) to get sbi pointer to
access .raw_super field, to avoid unneeded pointer conversion, this
patch changes to F2FS_HAS_FEATURE() accept sbi parameter directly.
Just do cleanup, no logic change.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
When sbi->segs_per_sec > 1, and if some segno has 0 valid blocks before
gc starts, do_garbage_collect will skip counting seg_freed++, and this
will cause seg_freed < sbi->segs_per_sec and finally skip sec_freed++.
Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Previously, we only account preflush command for flush_merge mode,
so for noflush_merge mode, we can not know in-flight preflush
command count, fix it.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
The encrypted file may be corrupted by GC in following case:
Time 1: | segment 1 blkaddr = A | GC -> | segment 2 blkaddr = B |
Encrypted block 1 is moved from blkaddr A of segment 1 to blkaddr B of
segment 2,
Time 2: | segment 1 blkaddr = B | GC -> | segment 3 blkaddr = C |
Before page 1 is written back and if segment 2 become a victim, then
page 1 is moved from blkaddr B of segment 2 to blkaddr Cof segment 3,
during the GC process of Time 2, f2fs should wait for page 1 written back
before reading it, or move_data_block will read a garbage block from
blkaddr B since page is not written back to blkaddr B yet.
Commit 6aa58d8a ("f2fs: readahead encrypted block during GC") introduce
ra_data_block to read encrypted block, but it forgets to add
f2fs_wait_on_page_writeback to avoid racing between GC and flush.
Signed-off-by: Yunlong Song <yunlong.song@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
For journalled quota mode, let checkpoint to flush dquot dirty data
and quota file data to guarntee persistence of all quota sysfile in
last checkpoint, by this way, we can avoid corrupting quota sysfile
when encountering SPO.
The implementation is as below:
1. add a global state SBI_QUOTA_NEED_FLUSH to indicate that there is
cached dquot metadata changes in quota subsystem, and later checkpoint
should:
a) flush dquot metadata into quota file.
b) flush quota file to storage to keep file usage be consistent.
2. add a global state SBI_QUOTA_NEED_REPAIR to indicate that quota
operation failed due to -EIO or -ENOSPC, so later,
a) checkpoint will skip syncing dquot metadata.
b) CP_QUOTA_NEED_FSCK_FLAG will be set in last cp pack to give a
hint for fsck repairing.
3. add a global state SBI_QUOTA_SKIP_FLUSH, in checkpoint, if quota
data updating is very heavy, it may cause hungtask in block_operation().
To avoid this, if our retry time exceed threshold, let's just skip
flushing and retry in next checkpoint().
Signed-off-by: Weichao Guo <guoweichao@huawei.com>
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: avoid warnings and set fsck flag]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Direct IO can be used in case of hardware encryption. The following
scenario results into data corruption issue in this path -
Thread A - Thread B-
-> write file#1 in direct IO
-> GC gets kicked in
-> GC submitted bio on meta mapping
for file#1, but pending completion
-> write file#1 again with new data
in direct IO
-> GC bio gets completed now
-> GC writes old data to the new
location and thus file#1 is
corrupted.
Fix this by submitting and waiting for pending io on meta mapping
for direct IO case in f2fs_map_blocks().
Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Testcase to reproduce this bug:
1. mkfs.f2fs /dev/sdd
2. mount -t f2fs /dev/sdd /mnt/f2fs
3. touch /mnt/f2fs/file
4. sync
5. chattr +a /mnt/f2fs/file
6. xfs_io -a /mnt/f2fs/file -c "fsync"
7. godown /mnt/f2fs
8. umount /mnt/f2fs
9. mount -t f2fs /dev/sdd /mnt/f2fs
10. xfs_io /mnt/f2fs/file
There is no error when opening this file w/o O_APPEND, but actually,
we expect the correct result should be:
/mnt/f2fs/file: Operation not permitted
The root cause is, in recover_inode(), we recover inode->i_flags more
than F2FS_I(inode)->i_flags, so fix it.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch changes codes as below:
- use f2fs_set_inode_flags() to update i_flags atomically to avoid
potential race.
- synchronize F2FS_I(inode)->i_flags to inode->i_flags in
f2fs_new_inode().
- use f2fs_set_inode_flags() to simply codes in f2fs_quota_{on,off}.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
We need to drop PG_checked flag on page as well when we clear PG_uptodate
flag, in order to avoid treating the page as GCing one later.
Signed-off-by: Weichao Guo <guoweichao@huawei.com>
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This reverts commit 66110abc4c931f879d70e83e1281f891699364bf.
If we clear the cold data flag out of the writeback flow, we can miscount
-1 by end_io, which incurs a deadlock caused by all I/Os being blocked during
heavy GC.
Balancing F2FS Async:
- IO (CP: 1, Data: -1, Flush: ( 0 0 1), Discard: ( ...
GC thread: IRQ
- move_data_page()
- set_page_dirty()
- clear_cold_data()
- f2fs_write_end_io()
- type = WB_DATA_TYPE(page);
here, we get wrong type
- dec_page_count(sbi, type);
- f2fs_wait_on_page_writeback()
Cc: <stable@vger.kernel.org>
Reported-and-Tested-by: Park Ju Hyung <qkrwngud825@gmail.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch adds issued read IO counts which is under block layer.
Chao modified a bit, since:
Below race can cause reversed reference on F2FS_RD_DATA, there is
the same issue in f2fs_submit_page_bio(), fix them by relocate
__submit_bio() and inc_page_count.
Thread A Thread B
- f2fs_write_begin
- f2fs_submit_page_read
- __submit_bio
- f2fs_read_end_io
- __read_end_io
- dec_page_count(, F2FS_RD_DATA)
- inc_page_count(, F2FS_RD_DATA)
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Now, we have supported cgroup writeback, it depends on correctly IO
account of specified filesystem.
But in commit d1b3e72d5490 ("f2fs: submit bio of in-place-update pages"),
we split write paths from f2fs_submit_page_mbio() to two:
- f2fs_submit_page_bio() for IPU path
- f2fs_submit_page_bio() for OPU path
But still we account write IO only in f2fs_submit_page_mbio(), result in
incorrect IO account, fix it by adding missing IO account in IPU path.
Fixes: d1b3e72d5490 ("f2fs: submit bio of in-place-update pages")
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Below race can cause reversed reference on dirty count, fix it by
relocating __submit_bio() and inc_page_count().
Thread A Thread B
- f2fs_inplace_write_data
- f2fs_submit_page_bio
- __submit_bio
- f2fs_write_end_io
- dec_page_count
- inc_page_count
Cc: <stable@vger.kernel.org>
Fixes: d1b3e72d5490 ("f2fs: submit bio of in-place-update pages")
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This doesn't work on stacked devices, and it doesn't work on
blk-mq devices. The request_list is only used on legacy, which
we don't have much of anymore, and soon won't have any of.
Kill the check.
Cc: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: linux-f2fs-devel@lists.sourceforge.net
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Since we can use the filesystem without quotas till next boot.
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
The REQ_TIME should be updated only in case of success cases
as followed at all other places in the file system.
Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Commit 7735730d39d7 ("f2fs: fix to propagate error from __get_meta_page()")
added disable_nat_bits() in error path of __get_nat_bitmaps(), but it's
unneeded, beause we will fail mount, we won't have chance to change nid
usage status w/o nat full/empty bitmaps.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Commit a2a12b679f36 ("f2fs: export SSR allocation threshold") introduced
two threshold .min_ssr_sections and .trigger_ssr_threshold, but only
.min_ssr_sections is used, so just remove redundant one for cleanup.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
file_set_{cold,hot} doesn't need holding sbi->sb_lock, so moving them
out of the lock.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Testcase to reproduce this bug:
1. mkfs.f2fs /dev/sdd
2. mount -t f2fs /dev/sdd /mnt/f2fs
3. touch /mnt/f2fs/file
4. sync
5. chattr +A /mnt/f2fs/file
6. xfs_io -f /mnt/f2fs/file -c "fsync"
7. godown /mnt/f2fs
8. umount /mnt/f2fs
9. mount -t f2fs /dev/sdd /mnt/f2fs
10. chattr -A /mnt/f2fs/file
11. xfs_io -f /mnt/f2fs/file -c "fsync"
12. umount /mnt/f2fs
13. mount -t f2fs /dev/sdd /mnt/f2fs
14. lsattr /mnt/f2fs/file
-----------------N- /mnt/f2fs/file
But actually, we expect the corrct result is:
-------A---------N- /mnt/f2fs/file
The reason is in step 9) we missed to recover cold bit flag in inode
block, so later, in fsync, we will skip write inode block due to below
condition check, result in lossing data in another SPOR.
f2fs_fsync_node_pages()
if (!IS_DNODE(page) || !is_cold_node(page))
continue;
Note that, I guess that some non-dir inode has already lost cold bit
during POR, so in order to reenable recovery for those inode, let's
try to recover cold bit in f2fs_iget() to save more fsynced data.
Fixes: c56675750d7c ("f2fs: remove unneeded set_cold_node()")
Cc: <stable@vger.kernel.org> 4.17+
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
When migrating encrypted block from background GC thread, we only add
them into f2fs inner bio cache, but forget to submit the cached bio, it
may cause potential deadlock when we are waiting page writebacked, fix
it.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Note that, it requires "f2fs: return correct errno in f2fs_gc".
This adds a lightweight non-persistent snapshotting scheme to f2fs.
To use, mount with the option checkpoint=disable, and to return to
normal operation, remount with checkpoint=enable. If the filesystem
is shut down before remounting with checkpoint=enable, it will revert
back to its apparent state when it was first mounted with
checkpoint=disable. This is useful for situations where you wish to be
able to roll back the state of the disk in case of some critical
failure.
Signed-off-by: Daniel Rosenberg <drosen@google.com>
[Jaegeuk Kim: use SB_RDONLY instead of MS_RDONLY]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
When running fault injection test, I hit somewhat wrong behavior in f2fs_gc ->
gc_data_segment():
0. fault injection generated some PageError'ed pages
1. gc_data_segment
-> f2fs_get_read_data_page(REQ_RAHEAD)
2. move_data_page
-> f2fs_get_lock_data_page()
-> f2f_get_read_data_page()
-> f2fs_submit_page_read()
-> submit_bio(READ)
-> return EIO due to PageError
-> fail to move data
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Normally, DIO uses in-pllace-update, but in LFS mode, f2fs doesn't
allow triggering any in-place-update writes, so we fallback direct
write to buffered write, result in bad performance of large size
write.
This patch adds to support triggering out-place-update for direct IO
to enhance its performance.
Note that it needs to exclude direct read IO during direct write,
since new data writing to new block address will no be valid until
write finished.
storage: zram
time xfs_io -f -d /mnt/f2fs/file -c "pwrite 0 1073741824" -c "fsync"
Before:
real 0m13.061s
user 0m0.327s
sys 0m12.486s
After:
real 0m6.448s
user 0m0.228s
sys 0m6.212s
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Thread A Thread B
- f2fs_vm_page_mkwrite
- f2fs_setattr
- down_write(i_mmap_sem)
- truncate_setsize
- f2fs_truncate
- up_write(i_mmap_sem)
- f2fs_reserve_block
reserve NEW_ADDR
- skip dirty page due to truncation
1. we don't need to rserve new block address for a truncated page.
2. dn.data_blkaddr is used out of node page lock coverage.
Refactor ->page_mkwrite() flow to fix above issues:
- use __do_map_lock() to avoid racing checkpoint()
- lock data page in prior to dnode page
- cover f2fs_reserve_block with i_mmap_sem lock
- wait page writeback before zeroing page
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
There is one case that we can leave bio in f2fs, result in hanging
page writeback waiter.
Thread A Thread B
- f2fs_write_cache_pages
- f2fs_submit_page_write
page #0 cached in bio #0 of cold log
- f2fs_submit_page_write
page #1 cached in bio #1 of warm log
- f2fs_write_cache_pages
- f2fs_submit_page_write
bio is full, submit bio #1 contain page #1
- f2fs_submit_merged_write_cond(, page #1)
fail to submit bio #0 due to page #1 is not in any cached bios.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Now we support crc32 checksum for superblock.
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Junling Zheng <zhengjunling@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch adds to account skip count of background GC, and show stat
info via 'status' debugfs entry.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch supports to account meta IO, it enables to show write IO
from f2fs more comprehensively via 'status' debugfs entry.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch fixes missing up_read call.
Fixes: c9b60788fc76 ("f2fs: fix to do sanity check with block address in main area")
Cc: <stable@vger.kernel.org> # 4.19+
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch avoids BUG_ON when f2fs_get_meta_page_nofail got EIO during
xfstests/generic/475.
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Mark inode dirty explicitly in the end of recover_inode() to make sure
that all recoverable fields can be persisted later.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
inode.i_gc_failures is used to indicate that skip count of migrating
on blocks of inode, we should guarantee it can be recovered in sudden
power-off case.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Testcase to reproduce this bug:
1. mkfs.f2fs /dev/sdd
2. mount -t f2fs /dev/sdd /mnt/f2fs
3. touch /mnt/f2fs/file
4. sync
5. chattr +A /mnt/f2fs/file
6. xfs_io -f /mnt/f2fs/file -c "fsync"
7. godown /mnt/f2fs
8. umount /mnt/f2fs
9. mount -t f2fs /dev/sdd /mnt/f2fs
10. lsattr /mnt/f2fs/file
-----------------N- /mnt/f2fs/file
But actually, we expect the corrct result is:
-------A---------N- /mnt/f2fs/file
The reason is we didn't recover inode.i_flags field during mount,
fix it.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
