We expect cold files write data sequentially, but sometimes some of small data
can be updated, which incurs fragmentation.
Let's avoid that.
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In this round, we've added new features such as disk quota and statx, and
modified internal bio management flow to merge more IOs depending on block
types. We've also made internal threads freezeable for Android battery life.
In addition to them, there are some patches to avoid lock contention as well
as a couple of deadlock conditions.
= Enhancement
- support usrquota, grpquota, and statx
- manage DATA/NODE typed bios separately to serialize more IOs
- modify f2fs_lock_op/wio_mutex to avoid lock contention
- prevent lock contention in migratepage
= Bug fix
- miss to load written inode flag
- fix worst case victim selection in GC
- freezeable GC and discard threads for Android battery life
- sanitize f2fs metadata to deal with security hole
- clean up sysfs-related code and docs
-----BEGIN PGP SIGNATURE-----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=N4GZ
-----END PGP SIGNATURE-----
Merge tag 'for-f2fs-4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs
Pull f2fs updates from Jaegeuk Kim:
"In this round, we've added new features such as disk quota and statx,
and modified internal bio management flow to merge more IOs depending
on block types. We've also made internal threads freezeable for
Android battery life. In addition to them, there are some patches to
avoid lock contention as well as a couple of deadlock conditions.
Enhancements:
- support usrquota, grpquota, and statx
- manage DATA/NODE typed bios separately to serialize more IOs
- modify f2fs_lock_op/wio_mutex to avoid lock contention
- prevent lock contention in migratepage
Bug fixes:
- fix missing load of written inode flag
- fix worst case victim selection in GC
- freezeable GC and discard threads for Android battery life
- sanitize f2fs metadata to deal with security hole
- clean up sysfs-related code and docs"
* tag 'for-f2fs-4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (59 commits)
f2fs: support plain user/group quota
f2fs: avoid deadlock caused by lock order of page and lock_op
f2fs: use spin_{,un}lock_irq{save,restore}
f2fs: relax migratepage for atomic written page
f2fs: don't count inode block in in-memory inode.i_blocks
Revert "f2fs: fix to clean previous mount option when remount_fs"
f2fs: do not set LOST_PINO for renamed dir
f2fs: do not set LOST_PINO for newly created dir
f2fs: skip ->writepages for {mete,node}_inode during recovery
f2fs: introduce __check_sit_bitmap
f2fs: stop gc/discard thread in prior during umount
f2fs: introduce reserved_blocks in sysfs
f2fs: avoid redundant f2fs_flush after remount
f2fs: report # of free inodes more precisely
f2fs: add ioctl to do gc with target block address
f2fs: don't need to check encrypted inode for partial truncation
f2fs: measure inode.i_blocks as generic filesystem
f2fs: set CP_TRIMMED_FLAG correctly
f2fs: require key for truncate(2) of encrypted file
f2fs: move sysfs code from super.c to fs/f2fs/sysfs.c
...
Split DATA/NODE type bio cache according to different temperature,
so write IOs with the same temperature can be merged in corresponding
bio cache as much as possible, otherwise, different temperature write
IOs submitting into one bio cache will always cause split of bio.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Merge more updates from Andrew Morton:
- the rest of MM
- various misc things
- procfs updates
- lib/ updates
- checkpatch updates
- kdump/kexec updates
- add kvmalloc helpers, use them
- time helper updates for Y2038 issues. We're almost ready to remove
current_fs_time() but that awaits a btrfs merge.
- add tracepoints to DAX
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (114 commits)
drivers/staging/ccree/ssi_hash.c: fix build with gcc-4.4.4
selftests/vm: add a test for virtual address range mapping
dax: add tracepoint to dax_insert_mapping()
dax: add tracepoint to dax_writeback_one()
dax: add tracepoints to dax_writeback_mapping_range()
dax: add tracepoints to dax_load_hole()
dax: add tracepoints to dax_pfn_mkwrite()
dax: add tracepoints to dax_iomap_pte_fault()
mtd: nand: nandsim: convert to memalloc_noreclaim_*()
treewide: convert PF_MEMALLOC manipulations to new helpers
mm: introduce memalloc_noreclaim_{save,restore}
mm: prevent potential recursive reclaim due to clearing PF_MEMALLOC
mm/huge_memory.c: deposit a pgtable for DAX PMD faults when required
mm/huge_memory.c: use zap_deposited_table() more
time: delete CURRENT_TIME_SEC and CURRENT_TIME
gfs2: replace CURRENT_TIME with current_time
apparmorfs: replace CURRENT_TIME with current_time()
lustre: replace CURRENT_TIME macro
fs: ubifs: replace CURRENT_TIME_SEC with current_time
fs: ufs: use ktime_get_real_ts64() for birthtime
...
CURRENT_TIME_SEC is not y2038 safe.
Replace use of CURRENT_TIME_SEC with ktime_get_real_seconds in segment
timestamps used by GC algorithm including the segment mtime timestamps.
Link: http://lkml.kernel.org/r/1491613030-11599-2-git-send-email-deepa.kernel@gmail.com
Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Cc: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch introduces encrypt_one_page which encrypts one data page before
submit_bio, and change the use of need_inplace_update.
Signed-off-by: Hou Pengyang <houpengyang@huawei.com>
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Async request may be throttled in block layer, so page for async may keep WRITE_BACK
for a long time.
For encrytped inode, we need wait on page writeback no matter if the device supports
BDI_CAP_STABLE_WRITES. This may result in a higher waiting page writeback time for
async encrypted inode page.
This patch skips IPU for encrypted inode's updating write.
Signed-off-by: Hou Pengyang <houpengyang@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch adds an ioctl to flush data in faster device to cold area. User can
give device number and number of segments to move. It doesn't move it if there
is only one device.
The parameter looks like:
struct f2fs_flush_device {
u32 dev_num; /* device number to flush */
u32 segments; /* # of segments to flush */
};
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch introduces an ASYNC IPU policy.
Under senario of large # of async updating(e.g. log writing in Android),
disk would be seriously fragmented, and higher frequent gc would be triggered.
This patch uses IPU to rewrite the async update writting, since async is
NOT sensitive to io latency.
Signed-off-by: Hou Pengyang <houpengyang@huawei.com>
Add braces around variables used within macros for those make sense
to do it. Many of the macros in f2fs already do this. What this commit
doesn't do is anything that changes line# as a result of adding braces,
which usually affects the binary via __LINE__.
Confirmed no diff in fs/f2fs/f2fs.ko before/after this commit on x86_64,
to make sure this has no functional change as well as there's been no
unexpected side effect due to callers' arithmetics within the existing
code.
Signed-off-by: Tomohiro Kusumi <tkusumi@tuxera.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
It would better split small and large IOs separately in order to get more
consecutive big writes.
The default threshold is set to 64KB, but configurable by sysfs/min_hot_blocks.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
For foreground gc, greedy algorithm should be adapted, which makes
this formula work well:
(2 * (100 / config.overprovision + 1) + 6)
But currently, we fg_gc have a prior to select bg_gc victim segments to gc
first, these victims are selected by cost-benefit algorithm, we can't guarantee
such segments have the small valid blocks, which may destroy the f2fs rule, on
the worstest case, would consume all the free segments.
This patch fix this by add a filter in check_bg_victims, if segment's has # of
valid blocks over overprovision ratio, skip such segments.
Cc: <stable@vger.kernel.org>
Signed-off-by: Hou Pengyang <houpengyang@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch adds a mirror for sit version bitmap, and use it to detect
in-memory bitmap corruption which may be caused by bit-transition of
cache or memory overflow.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch adds a mirror for valid block bitmap, and use it to detect
in-memory bitmap corruption which may be caused by bit-transition of
cache or memory overflow.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch implements IO alignment by filling dummy blocks in DATA and NODE
write bios. If we can guarantee, for example, 32KB or 64KB for such the IOs,
we can eliminate underlying dummy page problem which FTL conducts in order to
close MLC or TLC partial written pages.
Note that,
- it requires "-o mode=lfs".
- IO size should be power of 2, not exceed BIO_MAX_PAGES, 256.
- read IO is still 4KB.
- do checkpoint at fsync, if dummy NODE page was written.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Previous mkfs.f2fs allows small partition inappropriately, so f2fs should detect
that as well.
Refer this in f2fs-tools.
mkfs.f2fs: detect small partition by overprovision ratio and # of segments
Reported-and-Tested-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
If there are a lot of dirty inodes, we need to flush all of them when doing
checkpoint. So, we need to count this for enough free space.
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In 1TB storage, we need to admit 22841 prefree segments, which can consume
too much segments.
This patch sets 8GB in max. prefree segments in that case.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This mount option is to enable original log-structured filesystem forcefully.
So, there should be no random writes for main area.
Especially, this supports host-managed SMR device.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
For data pages, let's try to flush as much as possible in background.
On /dev/pmem0,
1. dd if=/dev/zero of=/mnt/test/testfile bs=1M count=2048 conv=fsync
Before : 800 MB/s
After : 1.1 GB/s
2. dd if=/dev/zero of=/mnt/test/testfile bs=1M count=2048
Before : 1.3 GB/s
After : 2.2 GB/s
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Restructure struct seg_entry to eliminate holes in it, after that,
in 32-bits machine, it reduces size from 32 bytes to 24 bytes; in
64-bits machine, it reduces size from 56 bytes to 40 bytes.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In curseg cache, f2fs caches two different parts:
- datas of current summay block, i.e. summary entries, footer info.
- journal info, i.e. sparse nat/sit entries or io stat info.
With this approach, 1) it may cause higher lock contention when we access
or update both of the parts of cache since we use the same mutex lock
curseg_mutex to protect the cache. 2) current summary block with last
journal info will be writebacked into device as a normal summary block
when flushing, however, we treat journal info as valid one only in current
summary, so most normal summary blocks contain junk journal data, it wastes
remaining space of summary block.
So, in order to fix above issues, we split curseg cache into two parts:
a) current summary block, protected by original mutex lock curseg_mutex
b) journal cache, protected by newly introduced r/w semaphore journal_rwsem
When loading curseg cache during ->mount, we store summary info and
journal info into different caches; When doing checkpoint, we combine
datas of two cache into current summary block for persisting.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
f2fs support atomic write with following semantics:
1. open db file
2. ioctl start atomic write
3. (write db file) * n
4. ioctl commit atomic write
5. close db file
With this flow we can avoid file becoming corrupted when abnormal power
cut, because we hold data of transaction in referenced pages linked in
inmem_pages list of inode, but without setting them dirty, so these data
won't be persisted unless we commit them in step 4.
But we should still hold journal db file in memory by using volatile
write, because our semantics of 'atomic write support' is incomplete, in
step 4, we could fail to submit all dirty data of transaction, once
partial dirty data was committed in storage, then after a checkpoint &
abnormal power-cut, db file will be corrupted forever.
So this patch tries to improve atomic write flow by adding a revoking flow,
once inner error occurs in committing, this gives another chance to try to
revoke these partial submitted data of current transaction, it makes
committing operation more like aotmical one.
If we're not lucky, once revoking operation was failed, EAGAIN will be
reported to user for suggesting doing the recovery with held journal file,
or retrying current transaction again.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
The periodic checkpoint can resolve the previous issue.
So, now we can use this again to improve the reported performance regression:
https://lkml.org/lkml/2015/10/8/20
This reverts commit 15bec0ff5a9ba6d203178fa8772259df6207942a.
Previously, we skip dentry block writes when wbc is SYNC_NONE with no memory
pressure and the number of dirty pages is pretty small.
But, we didn't skip for normal data writes, which gives us not much big impact
on overall performance.
Moreover, by skipping some data writes, kworker falls into infinite loop to try
to write blocks, when many dir inodes have only one dentry block.
So, this patch removes skipping data writes.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Previously, we use radix tree to index all registered page entries for
atomic file, but now we only use radix tree to see whether current page
is indexed or not, since the other user of radix tree is gone in commit
042b7816aa ("f2fs: remove unnecessary call to invalidate inmemory pages").
So in this patch, we try to use one more efficient way:
Introducing a macro ATOMIC_WRITTEN_PAGE, and setting it as page private
value to indicate page indexing status. By using this way, we can save
memory and lookup time.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Replace BUG_ON with f2fs_bug_on to deal with
block and segment validity check failed.
Signed-off-by: Xue Liu <liuxueliu.liu@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Pull cgroup writeback support from Jens Axboe:
"This is the big pull request for adding cgroup writeback support.
This code has been in development for a long time, and it has been
simmering in for-next for a good chunk of this cycle too. This is one
of those problems that has been talked about for at least half a
decade, finally there's a solution and code to go with it.
Also see last weeks writeup on LWN:
http://lwn.net/Articles/648292/"
* 'for-4.2/writeback' of git://git.kernel.dk/linux-block: (85 commits)
writeback, blkio: add documentation for cgroup writeback support
vfs, writeback: replace FS_CGROUP_WRITEBACK with SB_I_CGROUPWB
writeback: do foreign inode detection iff cgroup writeback is enabled
v9fs: fix error handling in v9fs_session_init()
bdi: fix wrong error return value in cgwb_create()
buffer: remove unusued 'ret' variable
writeback: disassociate inodes from dying bdi_writebacks
writeback: implement foreign cgroup inode bdi_writeback switching
writeback: add lockdep annotation to inode_to_wb()
writeback: use unlocked_inode_to_wb transaction in inode_congested()
writeback: implement unlocked_inode_to_wb transaction and use it for stat updates
writeback: implement [locked_]inode_to_wb_and_lock_list()
writeback: implement foreign cgroup inode detection
writeback: make writeback_control track the inode being written back
writeback: relocate wb[_try]_get(), wb_put(), inode_{attach|detach}_wb()
mm: vmscan: disable memcg direct reclaim stalling if cgroup writeback support is in use
writeback: implement memcg writeback domain based throttling
writeback: reset wb_domain->dirty_limit[_tstmp] when memcg domain size changes
writeback: implement memcg wb_domain
writeback: update wb_over_bg_thresh() to use wb_domain aware operations
...
With the planned cgroup writeback support, backing-dev related
declarations will be more widely used across block and cgroup;
unfortunately, including backing-dev.h from include/linux/blkdev.h
makes cyclic include dependency quite likely.
This patch separates out backing-dev-defs.h which only has the
essential definitions and updates blkdev.h to include it. c files
which need access to more backing-dev details now include
backing-dev.h directly. This takes backing-dev.h off the common
include dependency chain making it a lot easier to use it across block
and cgroup.
v2: fs/fat build failure fixed.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Jens Axboe <axboe@fb.com>
Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback)
and the role of the separation is unclear. For cgroup support for
writeback IOs, a bdi will be updated to host multiple wb's where each
wb serves writeback IOs of a different cgroup on the bdi. To achieve
that, a wb should carry all states necessary for servicing writeback
IOs for a cgroup independently.
This patch moves bandwidth related fields from backing_dev_info into
bdi_writeback.
* The moved fields are: bw_time_stamp, dirtied_stamp, written_stamp,
write_bandwidth, avg_write_bandwidth, dirty_ratelimit,
balanced_dirty_ratelimit, completions and dirty_exceeded.
* writeback_chunk_size() and over_bground_thresh() now take @wb
instead of @bdi.
* bdi_writeout_fraction(bdi, ...) -> wb_writeout_fraction(wb, ...)
bdi_dirty_limit(bdi, ...) -> wb_dirty_limit(wb, ...)
bdi_position_ration(bdi, ...) -> wb_position_ratio(wb, ...)
bdi_update_writebandwidth(bdi, ...) -> wb_update_write_bandwidth(wb, ...)
[__]bdi_update_bandwidth(bdi, ...) -> [__]wb_update_bandwidth(wb, ...)
bdi_{max|min}_pause(bdi, ...) -> wb_{max|min}_pause(wb, ...)
bdi_dirty_limits(bdi, ...) -> wb_dirty_limits(wb, ...)
* Init/exits of the relocated fields are moved to bdi_wb_init/exit()
respectively. Note that explicit zeroing is dropped in the process
as wb's are cleared in entirety anyway.
* As there's still only one bdi_writeback per backing_dev_info, all
uses of bdi->stat[] are mechanically replaced with bdi->wb.stat[]
introducing no behavior changes.
v2: Typo in description fixed as suggested by Jan.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
This patch adds a bitmap for discard issues from f2fs_trim_fs.
There-in rule is to issue discard commands only for invalidated blocks
after mount.
Once mount is done, f2fs_trim_fs trims out whole invalid area.
After ehn, it will not issue and discrads redundantly.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In __set_free we will check whether all segment are free in one section
when free one segment, in order to set section to free status. But the
searching region of segmap is from start segno to last segno of main
area, it's not necessary. So let's just only check all segment bitmap
of target section.
Signed-off-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
rwlock can provide better concurrency when there are much more readers than
writers because readers can hold the rwlock simultaneously.
But now, for segmap_lock rwlock in struct free_segmap_info, there is only one
reader 'mount' from below call path:
->f2fs_fill_super
->build_segment_manager
->build_dirty_segmap
->init_dirty_segmap
->find_next_inuse
read_lock
...
read_unlock
Now that our concurrency can not be improved since there is no other reader for
this lock, we do not need to use rwlock_t type for segmap_lock, let's replace it
with spinlock_t type.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Currently, there are several variables with Boolean type as below:
struct f2fs_sb_info {
...
int s_dirty;
bool need_fsck;
bool s_closing;
...
bool por_doing;
...
}
For this there are some issues:
1. there are some space of f2fs_sb_info is wasted due to aligning after Boolean
type variables by compiler.
2. if we continuously add new flag into f2fs_sb_info, structure will be messed
up.
So in this patch, we try to:
1. switch s_dirty to Boolean type variable since it has two status 0/1.
2. merge s_dirty/need_fsck/s_closing/por_doing variables into s_flag.
3. introduce an enum type which can indicate different states of sbi.
4. use new introduced universal interfaces is_sbi_flag_set/{set,clear}_sbi_flag
to operate flags for sbi.
After that, above issues will be fixed.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Introduce f2fs_change_bit to simplify the change bit logic in
function set_to_next_nat{sit}.
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch introduces a very limited functionality for atomic write support.
In order to support atomic write, this patch adds two ioctls:
o F2FS_IOC_START_ATOMIC_WRITE
o F2FS_IOC_COMMIT_ATOMIC_WRITE
The database engine should be aware of the following sequence.
1. open
-> ioctl(F2FS_IOC_START_ATOMIC_WRITE);
2. writes
: all the written data will be treated as atomic pages.
3. commit
-> ioctl(F2FS_IOC_COMMIT_ATOMIC_WRITE);
: this flushes all the data blocks to the disk, which will be shown all or
nothing by f2fs recovery procedure.
4. repeat to #2.
The IO pattens should be:
,- START_ATOMIC_WRITE ,- COMMIT_ATOMIC_WRITE
CP | D D D D D D | FSYNC | D D D D | FSYNC ...
`- COMMIT_ATOMIC_WRITE
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
My static checker complains that segment is a u64 but only the lower 31
bits can be used before we hit a shift wrapping bug.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Previously, f2fs activates SSR if the # of free segments reaches to the # of
overprovisioned segments.
In this case, SSR starts to use dirty segments only, so that the overprovisoned
space cannot be selected for new data.
This means that we have no chance to utilizae the overprovisioned space at all.
This patch fixes that by allowing LFS allocations until the # of free segments
reaches to the last threshold, reserved space.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch changes the ipu_policy setting to use any combination of orthogonal policies.
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Block size in f2fs is 4096 bytes, so theoretically, f2fs can support 4096 bytes
sector device at maximum. But now f2fs only support 512 bytes size sector, so
block device such as zRAM which uses page cache as its block storage space will
not be mounted successfully as mismatch between sector size of zRAM and sector
size of f2fs supported.
In this patch we support large sector size in f2fs, so block device with sector
size of 512/1024/2048/4096 bytes can be supported in f2fs.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Previously, all the dnode pages should be read during the roll-forward recovery.
Even worsely, whole the chain was traversed twice.
This patch removes that redundant and costly read operations by using page cache
of meta_inode and readahead function as well.
Reviewed-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
If user wrote F2FS_IPU_FSYNC:4 in /sys/fs/f2fs/ipu_policy, f2fs_sync_file
only starts to try in-place-updates.
And, if the number of dirty pages is over /sys/fs/f2fs/min_fsync_blocks, it
keeps out-of-order manner. Otherwise, it triggers in-place-updates.
This may be used by storage showing very high random write performance.
For example, it can be used when,
Seq. writes (Data) + wait + Seq. writes (Node)
is pretty much slower than,
Rand. writes (Data)
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In commit aec71382c6 ("f2fs: refactor flush_nat_entries codes for reducing NAT
writes"), we descripte the issue as below:
"Although building NAT journal in cursum reduce the read/write work for NAT
block, but previous design leave us lower performance when write checkpoint
frequently for these cases:
1. if journal in cursum has already full, it's a bit of waste that we flush all
nat entries to page for persistence, but not to cache any entries.
2. if journal in cursum is not full, we fill nat entries to journal util
journal is full, then flush the left dirty entries to disk without merge
journaled entries, so these journaled entries may be flushed to disk at next
checkpoint but lost chance to flushed last time."
Actually, we have the same problem in using SIT journal area.
In this patch, firstly we will update sit journal with dirty entries as many as
possible. Secondly if there is no space in sit journal, we will remove all
entries in journal and walk through the whole dirty entry bitmap of sit,
accounting dirty sit entries located in same SIT block to sit entry set. All
entry sets are linked to list sit_entry_set in sm_info, sorted ascending order
by count of entries in set. Later we flush entries in set which have fewest
entries into journal as many as we can, and then flush dense set with merged
entries to disk.
In this way we can use sit journal area more effectively, also we will reduce
SIT update, result in gaining in performance and saving lifetime of flash
device.
In my testing environment, it shows this patch can help to reduce SIT block
update obviously.
virtual machine + hard disk:
fsstress -p 20 -n 400 -l 5
sit page num cp count sit pages/cp
based 2006.50 1349.75 1.486
patched 1566.25 1463.25 1.070
Our latency of merging op is small when handling a great number of dirty SIT
entries in flush_sit_entries:
latency(ns) dirty sit count
36038 2151
49168 2123
37174 2232
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
sit_i in macro SIT_BLOCK_OFFSET/START_SEGNO is not used, remove it.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch replaces BUG cases with f2fs_bug_on to remain fsck.f2fs information.
And it implements some void functions to initiate fsck.f2fs too.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Fix typo and some grammatical errors.
The words "filesystem" and "readahead" are being used without the space treewide.
Signed-off-by: Park Ju Hyung <qkrwngud825@gmail.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch enforces in-place-updates only when fdatasync is requested.
If we adopt this in-place-updates for the fdatasync, we can skip to write the
recovery information.
Reviewed-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In __set_test_and_free we will check whether all segment are free in one section
When free one segment, in order to set section to free status.
But the searching region of segmap is from start segno to last segno of f2fs,
it's not necessary. So let's just only check all segment bitmap of target
section.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
It is more reasonable to determine the reclaiming rate of prefree segments
according to the volume size, which is set to 5% by default.
For example, if the volume is 128GB, the prefree segments are reclaimed
when the number reaches to 6.4GB.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch introduces nr_pages_to_write to align page writes to the segment
or other operational unit size, which can be tuned according to the system
environment.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch introduces nr_pages_to_skip(sbi, type) to determine writepages can
be skipped.
The dentry, node, and meta pages can be conrolled by F2FS without breaking the
FS consistency.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
The stat_show is just to show the current status of f2fs.
So, we can remove all the there-in locks.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Fixed a variety of trivial checkpatch warnings. The only delta should
be some minor formatting on log strings that were split / too long.
Signed-off-by: Chris Fries <cfries@motorola.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch introduces new sysfs entries for users to control the policy of
in-place-updates, namely IPU, in f2fs.
Sometimes f2fs suffers from performance degradation due to its out-of-place
update policy that produces many additional node block writes.
If the storage performance is very dependant on the amount of data writes
instead of IO patterns, we'd better drop this out-of-place update policy.
This patch suggests 5 polcies and their triggering conditions as follows.
[sysfs entry name = ipu_policy]
0: F2FS_IPU_FORCE all the time,
1: F2FS_IPU_SSR if SSR mode is activated,
2: F2FS_IPU_UTIL if FS utilization is over threashold,
3: F2FS_IPU_SSR_UTIL if SSR mode is activated and FS utilization is over
threashold,
4: F2FS_IPU_DISABLE disable IPU. (=default option)
[sysfs entry name = min_ipu_util]
This parameter controls the threshold to trigger in-place-updates.
The number indicates percentage of the filesystem utilization, and used by
F2FS_IPU_UTIL and F2FS_IPU_SSR_UTIL policies.
For more details, see need_inplace_update() in segment.h.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
As we know, some of our branch condition will rarely be true. So we could add
'unlikely' to let compiler optimize these code, by this way we could drop
unneeded 'jump' assemble code to improve performance.
change log:
o add *unlikely* as many as possible across the whole source files at once
suggested by Jaegeuk Kim.
Suggested-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Previously f2fs allocates its own bi_private data structure all the time even
though we don't use it. But, can we remove this bi_private allocation?
This patch removes such the additional bi_private allocation.
1. Retrieve f2fs_sb_info from its page->mapping->host->i_sb.
- This removes the usecases of bi_private in end_io.
2. Use bi_private only when we really need it.
- The bi_private is used only when the checkpoint procedure is conducted.
- When conducting the checkpoint, f2fs submits a META_FLUSH bio to wait its bio
completion.
- Since we have no dependancies to remove bi_private now, let's just use
bi_private pointer as the completion pointer.
Reviewed-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
The void *wait in bio_private is used for waiting completion of checkpoint bio.
So we don't need to use its type as void, but declare it as completion type.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
[Jaegeuk Kim: add description]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch fixes some bit overflows by the shift operations.
Dan Carpenter reported potential bugs on bit overflows as follows.
fs/f2fs/segment.c:910 submit_write_page()
warn: should 'blk_addr << ((sbi)->log_blocksize - 9)' be a 64 bit type?
fs/f2fs/checkpoint.c:429 get_valid_checkpoint()
warn: should '1 << ()' be a 64 bit type?
fs/f2fs/data.c:408 f2fs_readpage()
warn: should 'blk_addr << ((sbi)->log_blocksize - 9)' be a 64 bit type?
fs/f2fs/data.c:457 submit_read_page()
warn: should 'blk_addr << ((sbi)->log_blocksize - 9)' be a 64 bit type?
fs/f2fs/data.c:525 get_data_block_ro()
warn: should 'i << blkbits' be a 64 bit type?
Bug-Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Previously we read sit entries page one by one, this method lost the chance
of reading contiguous page together. So we read pages as contiguous as
possible for better mount performance.
change log:
o merge judgements/use 'Continue' or 'Break' instead of 'Goto' as Gu Zheng
suggested.
o add mark_page_accessed() before release page to delay VM reclaiming.
o remove '*order' for simplification of function as Jaegeuk Kim suggested.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
[Jaegeuk Kim: fix a bug on the block address calculation]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
It is not efficient comparing each segment type to find node or data.
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
[Jaegeuk Kim: remove unnecessary white spaces]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Previously, check_block_count check valid_map with bit data type in common
scenario that sit has all ones or zeros bitmap, it makes low mount performance.
So let's check the special bitmap with integer data type instead of the bit one.
v1-->v2:
o use find_next_{zero_}bit_le for better performance and readable as Jaegeuk
suggested.
o use neat logogram in comment as Gu Zheng suggested.
o search continuous ones or zeros for better performance when checking mixed
bitmap.
Suggested-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Signed-off-by: Shu Tan <shu.tan@samsung.com>
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
If you want to remove unnecessary BUG_ONs, you can just turn off F2FS_CHECK_FS
in your kernel config.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Previously, f2fs postpones reclaiming prefree segments into free segments
as much as possible.
However, if user writes and deletes a bunch of data without any sync or fsync
calls, some flash storages can suffer from garbage collections.
So, this patch adds the reclaiming codes to f2fs_write_node_pages and background
GC thread.
If there are a lot of prefree segments, let's do checkpoint so that f2fs
submits discard commands for the prefree regions to the flash storage.
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch add macro MAX_BIO_BLOCKS to limit value of npages in
f2fs_bio_alloc, it can avoid allocating failure in bio_alloc caused by
npages is larger than BIO_MAX_PAGES.
Signed-off-by: Yu Chao <chao2.yu@samsung.com>
Reviewed-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch improves the gc efficiency by optimizing the victim
selection policy. With this optimization, the random re-write
performance could increase up to 20%.
For f2fs, when disk is in shortage of free spaces, gc will selects
dirty segments and moves valid blocks around for making more space
available. The gc cost of a segment is determined by the valid blocks
in the segment. The less the valid blocks, the higher the efficiency.
The ideal victim segment is the one that has the most garbage blocks.
Currently, it searches up to 20 dirty segments for a victim segment.
The selected victim is not likely the best victim for gc when there
are much more dirty segments. Why not searching more dirty segments
for a better victim? The cost of searching dirty segments is
negligible in comparison to moving blocks.
In this patch, it enlarges the MAX_VICTIM_SEARCH to 4096 to make
the search more aggressively for a possible better victim. Since
it also applies to victim selection for SSR, it will likely improve
the SSR efficiency as well.
The test case is simple. It creates as many files until the disk full.
The size for each file is 32KB. Then it writes as many as 100000
records of 4KB size to random offsets of random files in sync mode.
The testing was done on a 2GB partition of a SDHC card. Let's see the
test result of f2fs without and with the patch.
---------------------------------------
2GB partition, SDHC
create 52023 files of size 32768 bytes
random re-write 100000 records of 4KB
---------------------------------------
| file creation (s) | rewrite time (s) | gc count | gc garbage blocks |
[no patch] 341 4227 1174 174840
[patched] 324 2958 645 106682
It's obvious that, with the patch, f2fs finishes the test in 20+% less
time than without the patch. And internally it does much less gc with
higher efficiency than before.
Since the performance improvement is related to gc, it might not be so
obvious for other tests that do not trigger gc as often as this one (
This is because f2fs selects dirty segments for SSR use most of the
time when free space is in shortage). The well-known iozone test tool
was not used for benchmarking the patch becuase it seems do not have
a test case that performs random re-write on a full disk.
This patch is the revised version based on the suggestion from
Jaegeuk Kim.
Signed-off-by: Jin Xu <jinuxstyle@gmail.com>
[Jaegeuk Kim: suggested simpler solution]
Reviewed-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
The current f2fs uses all the block counts with 32 bit numbers, which is able to
cover about 15TB volume.
But in calculation of utilization, f2fs multiplies the count by 100 which can
induce overflow.
This patch fixes this.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Previously, f2fs conducts SSR when free_sections() < overprovision_sections.
But, even though there are a lot of prefree segments, it can consider SSR only.
So, let's consider the number of prefree segments too for triggering SSR.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
When testing f2fs on an SSD, I found some 128 page IOs followed by 1 page IO
were issued by f2fs_write_node_pages.
This means that there were some mishandling flows which degrades performance.
Previous f2fs_write_node_pages determines the number of pages to be written,
nr_to_write, as follows.
1. The bio_get_nr_vecs returns 129 pages.
2. The bio_alloc makes a room for 128 pages.
3. The initial 128 pages go into one bio.
4. The existing bio is submitted, and a new bio is prepared for the last 1 page.
5. Finally, sync_node_pages submits the last 1 page bio.
The problem is from the use of bio_get_nr_vecs, so this patch replace it
with max_hw_blocks using queue_max_sectors.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Fix typo mistakes.
1. I think that it should be 'L' instead of 'V'.
2. and try to fix 'Front' instead of 'Frone'
Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Amit Sahrawat <a.sahrawat@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch removes a bitmap for victim segments selected by foreground GC, and
modifies the other bitmap for victim segments selected by background GC.
1) foreground GC bitmap
: We don't need to manage this, since we just only one previous victim section
number instead of the whole victim history.
The f2fs uses the victim section number in order not to allocate currently
GC'ed section to current active logs.
2) background GC bitmap
: This bitmap is used to avoid selecting victims repeatedly by background GCs.
In addition, the victims are able to be selected by foreground GCs, since
there is no need to read victim blocks during foreground GCs.
By the fact that the foreground GC reclaims segments in a section unit, it'd
be better to manage this bitmap based on the section granularity.
Reviewed-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Let's use a macro to get the total number of sections.
Reviewed-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
A macro should not use duplicate parameter names.
Reviewed-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Use div_u64 to fix overflow when calculating utilization.
*long int* is 4-bytes on 32-bit so (user blocks * 100) might be
overflow if disk size is over e.g. 512GB.
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Reviewed-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch makes clearer the ambiguous f2fs_gc flow as follows.
1. Remove intermediate checkpoint condition during f2fs_gc
(i.e., should_do_checkpoint() and GC_BLOCKED)
2. Remove unnecessary return values of f2fs_gc because of #1.
(i.e., GC_NODE, GC_OK, etc)
3. Simplify write_checkpoint() because of #2.
4. Clarify the main f2fs_gc flow.
o monitor how many freed sections during one iteration of do_garbage_collect().
o do GC more without checkpoints if we can't get enough free sections.
o do checkpoint once we've got enough free sections through forground GCs.
5. Adopt thread-logging (Slack-Space-Recycle) scheme more aggressively on data
log types. See. get_ssr_segement()
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Instead of evaluating the free_sections and then deciding to return
true/false from that path. We can directly use the evaluation condition
for returning proper value.
Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Amit Sahrawat <a.sahrawat@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Introduce accessor to get the sections based upon the block type
(node,dents...) and modify the functions : should_do_checkpoint,
has_not_enough_free_secs to use this accessor function to get
the node sections and dent sections.
Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Amit Sahrawat <a.sahrawat@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Practically, has_not_enough_free_secs() should calculate with the numbers of
current node and directory data blocks together.
Actually the equation was implemented in need_to_flush().
So, this patch removes need_flush() and moves the equation into
has_not_enough_free_secs().
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>