path: root/Documentation/filesystems/f2fs.txt
diff options
authorLinus Torvalds <torvalds@linux-foundation.org>2014-01-23 09:21:09 -0800
committerLinus Torvalds <torvalds@linux-foundation.org>2014-01-23 09:21:09 -0800
commit0d90d638720ba14874e34cbd8766e4dc3f14f458 (patch)
tree5080f0335c1454ae6a3ef6b29834af20ae08a952 /Documentation/filesystems/f2fs.txt
parent1d32bdafaaa8bcc4c39b41ab9f674887d147f188 (diff)
parentbf39c00a9a7f3cdb5ce7d6695d9f044daf8f0b53 (diff)
Merge tag 'for-f2fs-3.14' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs
Pull f2fs updates from Jaegeuk Kim: "In this round, a couple of sysfs entries were introduced to tune the f2fs at runtime. In addition, f2fs starts to support inline_data and improves the read/write performance in some workloads by refactoring bio-related flows. This patch-set includes the following major enhancement patches. - support inline_data - refactor bio operations such as merge operations and rw type assignment - enhance the direct IO path - enhance bio operations - truncate a node page when it becomes obsolete - add sysfs entries: small_discards, max_victim_search, and in-place-update - add a sysfs entry to control max_victim_search The other bug fixes are as follows. - fix a bug in truncate_partial_nodes - avoid warnings during sparse and build process - fix error handling flows - fix potential bit overflows And, there are a bunch of cleanups" * tag 'for-f2fs-3.14' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (95 commits) f2fs: drop obsolete node page when it is truncated f2fs: introduce NODE_MAPPING for code consistency f2fs: remove the orphan block page array f2fs: add help function META_MAPPING f2fs: move a branch for code redability f2fs: call mark_inode_dirty to flush dirty pages f2fs: clean checkpatch warnings f2fs: missing REQ_META and REQ_PRIO when sync_meta_pages(META_FLUSH) f2fs: avoid f2fs_balance_fs call during pageout f2fs: add delimiter to seperate name and value in debug phrase f2fs: use spinlock rather than mutex for better speed f2fs: move alloc new orphan node out of lock protection region f2fs: move grabing orphan pages out of protection region f2fs: remove the needless parameter of f2fs_wait_on_page_writeback f2fs: update documents and a MAINTAINERS entry f2fs: add a sysfs entry to control max_victim_search f2fs: improve write performance under frequent fsync calls f2fs: avoid to read inline data except first page f2fs: avoid to left uninitialized data in page when read inline data f2fs: fix truncate_partial_nodes bug ...
Diffstat (limited to 'Documentation/filesystems/f2fs.txt')
1 files changed, 24 insertions, 0 deletions
diff --git a/Documentation/filesystems/f2fs.txt b/Documentation/filesystems/f2fs.txt
index a3fe811bbdbc..b8d284975f0f 100644
--- a/Documentation/filesystems/f2fs.txt
+++ b/Documentation/filesystems/f2fs.txt
@@ -120,6 +120,8 @@ active_logs=%u Support configuring the number of active logs. In the
disable_ext_identify Disable the extension list configured by mkfs, so f2fs
does not aware of cold files such as media files.
inline_xattr Enable the inline xattrs feature.
+inline_data Enable the inline data feature: New created small(<~3.4k)
+ files can be written into inode block.
@@ -171,6 +173,28 @@ Files in /sys/fs/f2fs/<devname>
conduct checkpoint to reclaim the prefree segments
to free segments. By default, 100 segments, 200MB.
+ max_small_discards This parameter controls the number of discard
+ commands that consist small blocks less than 2MB.
+ The candidates to be discarded are cached until
+ checkpoint is triggered, and issued during the
+ checkpoint. By default, it is disabled with 0.
+ ipu_policy This parameter controls the policy of in-place
+ updates in f2fs. There are five policies:
+ 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.
+ max_victim_search This parameter controls the number of trials to
+ find a victim segment when conducting SSR and
+ cleaning operations. The default value is 4096
+ which covers 8GB block address range.

Privacy Policy