aops->write_begin may allocate a new page and make it visible only to have
mark_page_accessed called almost immediately after. Once the page is
visible the atomic operations are necessary which is noticable overhead
when writing to an in-memory filesystem like tmpfs but should also be
noticable with fast storage. The objective of the patch is to initialse
the accessed information with non-atomic operations before the page is
visible.
The bulk of filesystems directly or indirectly use
grab_cache_page_write_begin or find_or_create_page for the initial
allocation of a page cache page. This patch adds an init_page_accessed()
helper which behaves like the first call to mark_page_accessed() but may
called before the page is visible and can be done non-atomically.
The primary APIs of concern in this care are the following and are used
by most filesystems.
find_get_page
find_lock_page
find_or_create_page
grab_cache_page_nowait
grab_cache_page_write_begin
All of them are very similar in detail to the patch creates a core helper
pagecache_get_page() which takes a flags parameter that affects its
behavior such as whether the page should be marked accessed or not. Then
old API is preserved but is basically a thin wrapper around this core
function.
Each of the filesystems are then updated to avoid calling
mark_page_accessed when it is known that the VM interfaces have already
done the job. There is a slight snag in that the timing of the
mark_page_accessed() has now changed so in rare cases it's possible a page
gets to the end of the LRU as PageReferenced where as previously it might
have been repromoted. This is expected to be rare but it's worth the
filesystem people thinking about it in case they see a problem with the
timing change. It is also the case that some filesystems may be marking
pages accessed that previously did not but it makes sense that filesystems
have consistent behaviour in this regard.
The test case used to evaulate this is a simple dd of a large file done
multiple times with the file deleted on each iterations. The size of the
file is 1/10th physical memory to avoid dirty page balancing. In the
async case it will be possible that the workload completes without even
hitting the disk and will have variable results but highlight the impact
of mark_page_accessed for async IO. The sync results are expected to be
more stable. The exception is tmpfs where the normal case is for the "IO"
to not hit the disk.
The test machine was single socket and UMA to avoid any scheduling or NUMA
artifacts. Throughput and wall times are presented for sync IO, only wall
times are shown for async as the granularity reported by dd and the
variability is unsuitable for comparison. As async results were variable
do to writback timings, I'm only reporting the maximum figures. The sync
results were stable enough to make the mean and stddev uninteresting.
The performance results are reported based on a run with no profiling.
Profile data is based on a separate run with oprofile running.
async dd
3.15.0-rc3 3.15.0-rc3
vanilla accessed-v2
ext3 Max elapsed 13.9900 ( 0.00%) 11.5900 ( 17.16%)
tmpfs Max elapsed 0.5100 ( 0.00%) 0.4900 ( 3.92%)
btrfs Max elapsed 12.8100 ( 0.00%) 12.7800 ( 0.23%)
ext4 Max elapsed 18.6000 ( 0.00%) 13.3400 ( 28.28%)
xfs Max elapsed 12.5600 ( 0.00%) 2.0900 ( 83.36%)
The XFS figure is a bit strange as it managed to avoid a worst case by
sheer luck but the average figures looked reasonable.
samples percentage
ext3 86107 0.9783 vmlinux-3.15.0-rc4-vanilla mark_page_accessed
ext3 23833 0.2710 vmlinux-3.15.0-rc4-accessed-v3r25 mark_page_accessed
ext3 5036 0.0573 vmlinux-3.15.0-rc4-accessed-v3r25 init_page_accessed
ext4 64566 0.8961 vmlinux-3.15.0-rc4-vanilla mark_page_accessed
ext4 5322 0.0713 vmlinux-3.15.0-rc4-accessed-v3r25 mark_page_accessed
ext4 2869 0.0384 vmlinux-3.15.0-rc4-accessed-v3r25 init_page_accessed
xfs 62126 1.7675 vmlinux-3.15.0-rc4-vanilla mark_page_accessed
xfs 1904 0.0554 vmlinux-3.15.0-rc4-accessed-v3r25 init_page_accessed
xfs 103 0.0030 vmlinux-3.15.0-rc4-accessed-v3r25 mark_page_accessed
btrfs 10655 0.1338 vmlinux-3.15.0-rc4-vanilla mark_page_accessed
btrfs 2020 0.0273 vmlinux-3.15.0-rc4-accessed-v3r25 init_page_accessed
btrfs 587 0.0079 vmlinux-3.15.0-rc4-accessed-v3r25 mark_page_accessed
tmpfs 59562 3.2628 vmlinux-3.15.0-rc4-vanilla mark_page_accessed
tmpfs 1210 0.0696 vmlinux-3.15.0-rc4-accessed-v3r25 init_page_accessed
tmpfs 94 0.0054 vmlinux-3.15.0-rc4-accessed-v3r25 mark_page_accessed
[akpm@linux-foundation.org: don't run init_page_accessed() against an uninitialised pointer]
Signed-off-by: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jan Kara <jack@suse.cz>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Tested-by: Prabhakar Lad <prabhakar.csengg@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch use list_for_each_entry{_safe} instead of list_for_each{_safe} for
simplfying code.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Move kmem_cache_free out of spinlock protection region for better performance.
Change log from v1:
o remove spinlock protection for kmem_cache_free in destroy_node_manager
suggested by Jaegeuk Kim.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
During the cleaing of node segments, F2FS can get errored node blocks due to
data race between node page lock and its valid bitmap operations.
In that case, it needs to return an error to skip such the obsolete block copy.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
If multiple redundant fsync calls are triggered, we don't need to write its
node pages with fsync mark continuously.
So, this patch adds FI_NEED_FSYNC to track whether the latest node block is
written with the fsync mark or not.
If the mark was set, a new fsync doesn't need to write a node block.
Otherwise, we should do a new node block with the mark for roll-forward
recovery.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
The NM_WOUT_THRESHOLD is now obsolete since f2fs starts to control on a basis
of the memory footprint.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch introduces ram_thresh, a sysfs entry, which controls the memory
footprint used by the free nid list and the nat cache.
Previously, the free nid list was controlled by MAX_FREE_NIDS, while the nat
cache was managed by NM_WOUT_THRESHOLD.
However, this approach cannot be applied dynamically according to the system.
So, this patch adds ram_thresh that users can specify the threshold, which is
in order of 1 / 1024.
For example, if the total ram size is 4GB and the value is set to 10 by default,
f2fs tries to control the number of free nids and nat caches not to consume over
10 * (4GB / 1024) = 10MB.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
The try_to_free_nats should not receive the negative nr_shrink.
Otherwise, it can drop all the nat entries by the while loop.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
If a page is on writeback, f2fs can face with deadlock due to under writepages.
This is caused by merging IOs inside f2fs, so if it comes to detect, let's throw
merged IOs, which is implemented by f2fs_wait_on_page_writeback.
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>
This patch introduces a help function f2fs_has_xattr_block for better
readability.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch introduces a help function f2fs_has_inline_xattr for better
readability.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Previously we do not recover inline xattr data of inode after power-cut, so
inline xattr data may be lost.
We should recover the data during the roll-forward process.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Previously, we ra_sum_pages to pre-read contiguous pages as more
as possible, and if we fail to alloc more pages, an ENOMEM error
will be reported upstream, even though we have alloced some pages
yet. In fact, we can use the available pages to do the job partly,
and continue the rest in the following circle. Only reporting ENOMEM
upstream if we really can not alloc any available page.
And another fix is ignoring dealing with the following pages if an
EIO occurs when reading page from page_list.
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Reviewed-by: Chao Yu <chao2.yu@samsung.com>
[Jaegeuk Kim: modify the flow for better neat code]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Integrated a couple of minor changes for better readability suggested by
Chao Yu.
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Reviewed-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
We should de-account dirty counters for page when redirty in ->writepage().
Wu Fengguang described in 'commit 971767caf632190f77a40b4011c19948232eed75':
"writeback: fix dirtied pages accounting on redirty
De-account the accumulative dirty counters on page redirty.
Page redirties (very common in ext4) will introduce mismatch between
counters (a) and (b)
a) NR_DIRTIED, BDI_DIRTIED, tsk->nr_dirtied
b) NR_WRITTEN, BDI_WRITTEN
This will introduce systematic errors in balanced_rate and result in
dirty page position errors (ie. the dirty pages are no longer balanced
around the global/bdi setpoints)."
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch introduces a radix tree for the list of free_nids, which enhances
the performance on free nid management.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Introduce help macro on_build_free_nids() which just uses build_lock
to judge whether the building free nid is going, so that we can remove
the on_build_free_nids field from f2fs_sb_info.
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
[Jaegeuk Kim: remove an unnecessary white line removal]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
The nat cache entry maintains a status whether it is checkpointed or not.
So, if a new cache entry is loaded from the last checkpoint,
nat_entry->checkpointed should be true.
If the cache entry is modified as being dirty, nat_entry->checkpoint should
be false.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch help us to cleanup the readahead code by merging ra_{sit,nat}_pages
function into ra_meta_pages.
Additionally the new function is used to readahead cp block in
recover_orphan_inodes.
Change log from v1:
o fix a deadloop bug pointed by Jaegeuk Kim.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
If a new xattr node page was allocated and its inode is fsynced, we should
recover the xattr node page during the roll-forward process after power-cut.
But, previously, f2fs didn't handle that case, resulting in kernel panic as
follows reported by Tom Li.
BUG: unable to handle kernel paging request at ffffc9001c861a98
IP: [<ffffffffa0295236>] check_index_in_prev_nodes+0x86/0x2d0 [f2fs]
Call Trace:
[<ffffffff815ece9b>] ? printk+0x48/0x4a
[<ffffffffa029626a>] recover_fsync_data+0xdca/0xf50 [f2fs]
[<ffffffffa02873ae>] f2fs_fill_super+0x92e/0x970 [f2fs]
[<ffffffff8112c9f8>] mount_bdev+0x1b8/0x200
[<ffffffffa0286a80>] ? f2fs_remount+0x130/0x130 [f2fs]
[<ffffffffa0285e40>] f2fs_mount+0x10/0x20 [f2fs]
[<ffffffff8112d4de>] mount_fs+0x3e/0x1b0
[<ffffffff810ef4eb>] ? __alloc_percpu+0xb/0x10
[<ffffffff8114761f>] vfs_kern_mount+0x6f/0x120
[<ffffffff811497b9>] do_mount+0x259/0xa90
[<ffffffff810ead1d>] ? memdup_user+0x3d/0x80
[<ffffffff810eadb3>] ? strndup_user+0x53/0x70
[<ffffffff8114a2c9>] SyS_mount+0x89/0xd0
[<ffffffff815feae2>] system_call_fastpath+0x16/0x1b
This patch adds a recovery function of xattr node pages.
Reported-by: Tom Li <biergaizi@members.fsf.org>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
If a node page is trucated, we'd better drop the page in the node_inode's page
cache for better memory footprint.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch adds NODE_MAPPING which is similar as META_MAPPING introduced by
Gu Zheng.
Cc: Gu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Introduce help function META_MAPPING() to get the cache meta blocks'
address space.
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
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>
When considering a bunch of data writes with very frequent fsync calls, we
are able to think the following performance regression.
N: Node IO, D: Data IO, IO scheduler: cfq
Issue pending IOs
D1 D2 D3 D4
D1 D2 D3 D4 N1
D2 D3 D4 N1 N2
N1 D3 D4 N2 D1
--> N1 can be selected by cfq becase of the same priority of N and D.
Then D3 and D4 would be delayed, resuling in performance degradation.
So, when processing the fsync call, it'd better give higher priority to data IOs
than node IOs by assigning WRITE and WRITE_SYNC respectively.
This patch improves the random wirte performance with frequent fsync calls by up
to 10%.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
The truncate_partial_nodes puts pages incorrectly in the following two cases.
Note that the value for argc 'depth' can only be 2 or 3.
Please see truncate_inode_blocks() and truncate_partial_nodes().
1) An err is occurred in the first 'for' loop
When err is occurred with depth = 2, pages[0] is invalid, so this page doesn't
need to be put. There is no problem, however, when depth is 3, it doesn't put
the pages correctly where pages[0] is valid and pages[1] is invalid.
In this case, depth is set to 2 (ref to statemnt depth = i + 1), and then
'goto fail'.
In label 'fail', for (i = depth - 3; i >= 0; i--) cannot meet the condition
because i = -1, so pages[0] cann't be put.
2) An err happened in the second 'for' loop
Now we've got pages[0] with depth = 2, or we've got pages[0] and pages[1]
with depth = 3. When an err is detected, we need 'goto fail' to put such
the pages.
When depth is 2, in label 'fail', for (i = depth - 3; i >= 0; i--) cann't
meet the condition because i = -1, so pages[0] cann't be put.
When depth is 3, in label 'fail', for (i = depth - 3; i >= 0; i--) can
only put pages[0], pages[1] also cann't be put.
Note that 'depth' has been changed before first 'goto fail' (ref to statemnt
depth = i + 1), so passing this modified 'depth' to the tracepoint,
trace_f2fs_truncate_partial_nodes, is also incorrect.
Signed-off-by: Shifei Ge <shifei10.ge@samsung.com>
[Jaegeuk Kim: modify the description and fix one bug]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch introduces F2FS_INODE that returns struct f2fs_inode * from the inode
page.
By using this macro, we can remove unnecessary casting codes like below.
struct f2fs_inode *ri = &F2FS_NODE(inode_page)->i;
-> struct f2fs_inode *ri = F2FS_INODE(inode_page);
Reviewed-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Update several comments:
1. use f2fs_{un}lock_op install of mutex_{un}lock_op.
2. update comment of get_data_block().
3. update description of node offset.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
When using the f2fs_io_info in the low level, we still need to merge the
rw and rw_flag, so use the rw to hold all the io flags directly,
and remove the rw_flag field.
ps.It is based on the previous patch:
f2fs: move all the bio initialization into __bio_alloc
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch introduces f2fs_io_info to mitigate the complex parameter list.
struct f2fs_io_info {
enum page_type type; /* contains DATA/NODE/META/META_FLUSH */
int rw; /* contains R/RS/W/WS */
int rw_flag; /* contains REQ_META/REQ_PRIO */
}
1. f2fs_write_data_pages
- DATA
- WRITE_SYNC is set when wbc->WB_SYNC_ALL.
2. sync_node_pages
- NODE
- WRITE_SYNC all the time
3. sync_meta_pages
- META
- WRITE_SYNC all the time
- REQ_META | REQ_PRIO all the time
** f2fs_submit_merged_bio() handles META_FLUSH.
4. ra_nat_pages, ra_sit_pages, ra_sum_pages
- META
- READ_SYNC
Cc: Fan Li <fanofcode.li@samsung.com>
Cc: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch adds unlikely() macro into the most of codes.
The basic rule is to add that when:
- checking unusual errors,
- checking page mappings,
- and the other unlikely conditions.
Change log from v1:
- Don't add unlikely for the NULL test and error test: advised by Andi Kleen.
Cc: Chao Yu <chao2.yu@samsung.com>
Cc: Andi Kleen <andi@firstfloor.org>
Reviewed-by: Chao Yu <chao2.yu@samsung.com>
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>
Use inner macro GFP_F2FS_ZERO to instead of GFP_NOFS | __GFP_ZERO for
simplification of code.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
If cp has no CP_UMOUNT_FLAG, we will read all pages in whole node segment
one by one, it makes low performance. So let's merge contiguous pages and
readahead for better performance.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
[Jaegeuk Kim: adjust the new bio operations]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch integrates redundant bio operations on read and write IOs.
1. Move bio-related codes to the top of data.c.
2. Replace f2fs_submit_bio with f2fs_submit_merged_bio, which handles read
bios additionally.
3. Introduce __submit_merged_bio to submit the merged bio.
4. Change f2fs_readpage to f2fs_submit_page_bio.
5. Introduce f2fs_submit_page_mbio to integrate previous submit_read_page and
submit_write_page.
Reviewed-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Reviewed-by: Chao Yu <chao2.yu@samsung.com >
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
The inode_page_locked should be a boolean variable.
struct dnode_of_data {
struct inode *inode; /* vfs inode pointer */
struct page *inode_page; /* its inode page, NULL is possible */
struct page *node_page; /* cached direct node page */
nid_t nid; /* node id of the direct node block */
unsigned int ofs_in_node; /* data offset in the node page */
==> bool inode_page_locked; /* inode page is locked or not */
block_t data_blkaddr; /* block address of the node block */
};
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
[Jaegeuk Kim: add description]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Let's send REQ_META or REQ_PRIO when reading meta area such as NAT/SIT
etc.
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Change log from v1:
o add mark_page_accessed() not to reclaim the nat pages.
This patch changes the policy of submitting read bios at ra_nat_pages.
Previously, f2fs submits small read bios with block plugging.
But, with this patch, f2fs itself merges read bios first and then submits a
large bio, which can reduce the bio handling overheads.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
In wait_on_node_pages_writeback we will test and clear error flag for all
pages in radix tree, but not necessary.
So we only do this for pages belong to the specified inode.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Previously, f2fs_sync_file() waits for all the node blocks to be written.
But, we don't need to do that, but wait only the inode-related node blocks.
This patch adds wait_on_node_pages_writeback() in which waits inode-related
node blocks that are on writeback.
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>
This patch merges some background jobs into this new function.
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Mel Gorman