diff options
| author | Jaegeuk Kim <jaegeuk@kernel.org> | 2016-12-14 10:12:56 -0800 |
|---|---|---|
| committer | Jaegeuk Kim <jaegeuk@kernel.org> | 2017-01-29 12:46:01 +0900 |
| commit | 0a595ebaaa6b53a2226d3fee2a2fd616ea5ba378 (patch) | |
| tree | 1479759e90462e5acd33f067bd9b41c1fac81b7a /include/linux/f2fs_fs.h | |
| parent | 554b5125f5cfca6653461fd52bad24d4ef35ec29 (diff) | |
| download | linux-stable-0a595ebaaa6b53a2226d3fee2a2fd616ea5ba378.tar.gz linux-stable-0a595ebaaa6b53a2226d3fee2a2fd616ea5ba378.tar.bz2 linux-stable-0a595ebaaa6b53a2226d3fee2a2fd616ea5ba378.zip | |
f2fs: support IO alignment for DATA and NODE writes
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>
Diffstat (limited to 'include/linux/f2fs_fs.h')
| -rw-r--r-- | include/linux/f2fs_fs.h | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h index cea41a124a80..f0748524ca8c 100644 --- a/include/linux/f2fs_fs.h +++ b/include/linux/f2fs_fs.h @@ -36,6 +36,12 @@ #define F2FS_NODE_INO(sbi) (sbi->node_ino_num) #define F2FS_META_INO(sbi) (sbi->meta_ino_num) +#define F2FS_IO_SIZE(sbi) (1 << (sbi)->write_io_size_bits) /* Blocks */ +#define F2FS_IO_SIZE_KB(sbi) (1 << ((sbi)->write_io_size_bits + 2)) /* KB */ +#define F2FS_IO_SIZE_BYTES(sbi) (1 << ((sbi)->write_io_size_bits + 12)) /* B */ +#define F2FS_IO_SIZE_BITS(sbi) ((sbi)->write_io_size_bits) /* power of 2 */ +#define F2FS_IO_SIZE_MASK(sbi) (F2FS_IO_SIZE(sbi) - 1) + /* This flag is used by node and meta inodes, and by recovery */ #define GFP_F2FS_ZERO (GFP_NOFS | __GFP_ZERO) #define GFP_F2FS_HIGH_ZERO (GFP_NOFS | __GFP_ZERO | __GFP_HIGHMEM) |