This adds a header file describing the on-disk layout of f2fs.
Signed-off-by: Changman Lee <[email protected]>
Signed-off-by: Chul Lee <[email protected]>
Signed-off-by: Jaegeuk Kim <[email protected]>
---
include/linux/f2fs_fs.h | 359 +++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 359 insertions(+)
create mode 100644 include/linux/f2fs_fs.h
diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h
new file mode 100644
index 0000000..b17eeec
--- /dev/null
+++ b/include/linux/f2fs_fs.h
@@ -0,0 +1,359 @@
+/**
+ * include/linux/f2fs_fs.h
+ *
+ * Copyright (c) 2012 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef _LINUX_F2FS_FS_H
+#define _LINUX_F2FS_FS_H
+
+#include <linux/pagemap.h>
+#include <linux/types.h>
+
+#define F2FS_SUPER_MAGIC 0xF2F52010
+#define F2FS_SUPER_OFFSET 1 /* start sector # for sb */
+#define F2FS_BLKSIZE 4096
+
+#define NULL_ADDR 0x0U
+#define NEW_ADDR -1U
+
+#define F2FS_ROOT_INO(sbi) (sbi->root_ino_num)
+#define F2FS_NODE_INO(sbi) (sbi->node_ino_num)
+#define F2FS_META_INO(sbi) (sbi->meta_ino_num)
+
+#define GFP_F2FS_MOVABLE (__GFP_WAIT | __GFP_IO | __GFP_ZERO)
+
+/*
+ * For superblock
+ */
+struct f2fs_super_block {
+ __le32 magic; /* Magic Number */
+ __le16 major_ver; /* Major Version */
+ __le16 minor_ver; /* Minor Version */
+ __le32 log_sectorsize; /* log2 (Sector size in bytes) */
+ __le32 log_sectors_per_block; /* log2 (Number of sectors per block */
+ __le32 log_blocksize; /* log2 (Block size in bytes) */
+ __le32 log_blocks_per_seg; /* log2 (Number of blocks per segment) */
+ __le32 log_segs_per_sec; /* log2 (Number of segments per section) */
+ __le32 secs_per_zone; /* Number of sections per zone */
+ __le32 checksum_offset; /* Checksum position in this super block */
+ __le64 block_count; /* Total number of blocks */
+ __le32 section_count; /* Total number of sections */
+ __le32 segment_count; /* Total number of segments */
+ __le32 segment_count_ckpt; /* Total number of segments
+ in Checkpoint area */
+ __le32 segment_count_sit; /* Total number of segments
+ in Segment information table */
+ __le32 segment_count_nat; /* Total number of segments
+ in Node address table */
+ /*Total number of segments in Segment summary area */
+ __le32 segment_count_ssa;
+ /* Total number of segments in Main area */
+ __le32 segment_count_main;
+ __le32 failure_safe_block_distance;
+ __le64 segment0_blkaddr; /* Start block address of Segment 0 */
+ __le64 start_segment_checkpoint; /* Start block address of ckpt */
+ __le64 sit_blkaddr; /* Start block address of SIT */
+ __le64 nat_blkaddr; /* Start block address of NAT */
+ __le64 ssa_blkaddr; /* Start block address of SSA */
+ __le64 main_blkaddr; /* Start block address of Main area */
+ __le32 root_ino; /* Root directory inode number */
+ __le32 node_ino; /* node inode number */
+ __le32 meta_ino; /* meta inode number */
+ __le32 volume_serial_number; /* VSN is optional field */
+ __le16 volume_name[8]; /* Volume Name. 8 unicode characters */
+} __packed;
+
+/*
+ * For checkpoint
+ */
+struct f2fs_checkpoint {
+ __le64 checkpoint_ver; /* Checkpoint block version number */
+ __le64 user_block_count; /* Total number of blocks
+ in Main area excluding the number of
+ reserved blocks */
+ __le64 valid_block_count; /* Total number of valid blocks
+ in Main area */
+ __le32 rsvd_segment_count; /* Total number of reserved segments
+ (for garbage collection) */
+ __le32 overprov_segment_count; /* Total number of overprovision
+ segments */
+ __le32 free_segment_count; /* Total number of free segments
+ in Main area */
+ __le32 bad_segment_count; /* Total number of bad segments
+ in Main area */
+ __le32 cur_node_segno[3]; /* Segment number of current node
+ segment */
+ __le16 cur_node_blkoff[3]; /* Current node block offset
+ in the node segment */
+ __le16 nat_upd_blkoff[3]; /* Block offset in current node segment
+ where the last NAT update happened */
+ __le32 cur_data_segno[3]; /* Segment number of current log
+ segment */
+ __le16 cur_data_blkoff[3]; /* Current data block offset
+ in the data segment */
+ __le32 ckpt_flags; /* Flags : umount and orphan_present */
+ __le32 cp_pack_total_block_count;
+ __le32 cp_pack_start_sum; /* start block number of data summary */
+ __le32 valid_node_count; /* Total number of valid nodes */
+ __le32 valid_inode_count; /* Total number of valid inodes */
+ __le32 next_free_nid; /* Next free node number */
+ __le32 sit_ver_bitmap_bytesize; /* Default value 64 */
+ __le32 nat_ver_bitmap_bytesize; /* Default value 256 */
+ __le32 checksum_offset; /* Checksum position
+ in this checkpoint block */
+ __le64 elapsed_time; /* elapsed time while partition
+ is mounted */
+ unsigned char alloc_type[6]; /* allocation type of current segment */
+
+ /* SIT and NAT version bitmap */
+ unsigned char sit_nat_version_bitmap[1];
+} __packed;
+
+/*
+ * For orphan inode management
+ */
+#define F2FS_ORPHANS_PER_BLOCK 1020
+
+struct f2fs_orphan_block {
+ __le32 ino[F2FS_ORPHANS_PER_BLOCK]; /* inode numbers */
+ __le32 reserved;
+ __le16 blk_addr; /* block index in current CP */
+ __le16 blk_count; /* Number of orphan inode blocks in CP */
+ __le32 entry_count; /* Total number of orphan nodes in current CP */
+ __le32 check_sum; /* CRC32 for orphan inode block */
+} __packed;
+
+/*
+ * For NODE structure
+ */
+struct f2fs_extent {
+ __le32 fofs;
+ __le32 blk_addr;
+ __le32 len;
+} __packed;
+
+#define F2FS_MAX_NAME_LEN 256
+#define ADDRS_PER_INODE 929 /* Address Pointers in an Inode */
+#define ADDRS_PER_BLOCK 1018 /* Address Pointers in a Direct Block */
+#define NIDS_PER_BLOCK 1018 /* Node IDs in an Indirect Block */
+
+struct f2fs_inode {
+ __le16 i_mode; /* File mode */
+ __le16 i_reserved; /* Reserved */
+ __le32 i_uid; /* User ID */
+ __le32 i_gid; /* Group ID */
+ __le32 i_links; /* Links count */
+ __le64 i_size; /* File size in bytes */
+ __le64 i_blocks; /* File size in bytes */
+ __le32 i_atime; /* Access time */
+ __le32 i_ctime; /* inode Change time */
+ __le32 i_mtime; /* Modification time */
+ __le32 i_btime; /* file creation time*/
+ __le32 current_depth;
+ __le32 i_xattr_nid; /* nid to save xattr */
+ __le32 i_flags; /* file attributes */
+ __le32 i_pino; /* parent inode number */
+ __le32 i_namelen; /* file name length */
+ __u8 i_name[F2FS_MAX_NAME_LEN]; /* file name for SPOR */
+
+ struct f2fs_extent i_ext; /* caching a largest extent */
+
+ __le32 i_addr[ADDRS_PER_INODE]; /* Pointers to data blocks */
+
+ __le32 i_nid[5]; /* direct(2), indirect(2),
+ double_indirect(1) node id */
+} __packed;
+
+struct direct_node {
+ __le32 addr[ADDRS_PER_BLOCK]; /* aray of data block address */
+} __packed;
+
+struct indirect_node {
+ __le32 nid[NIDS_PER_BLOCK]; /* aray of data block address */
+} __packed;
+
+struct node_footer {
+ __le32 nid; /* node id */
+ __le32 ino; /* inode nunmber */
+ __le32 cold:1; /* cold mark */
+ __le32 fsync:1; /* fsync mark */
+ __le32 dentry:1; /* dentry mark */
+ __le32 offset:29; /* offset in inode's node space */
+ __le64 cp_ver; /* checkpoint version */
+ __le32 next_blkaddr; /* next node page block address */
+} __packed;
+
+struct f2fs_node {
+ union {
+ struct f2fs_inode i;
+ struct direct_node dn;
+ struct indirect_node in;
+ };
+ struct node_footer footer;
+} __packed;
+
+/*
+ * For NAT entries
+ */
+#define NAT_ENTRY_PER_BLOCK (PAGE_CACHE_SIZE / sizeof(struct f2fs_nat_entry))
+
+struct f2fs_nat_entry {
+ __u8 version;
+ __le32 ino;
+ __le32 block_addr;
+} __packed;
+
+struct f2fs_nat_block {
+ struct f2fs_nat_entry entries[NAT_ENTRY_PER_BLOCK];
+} __packed;
+
+/*
+ * For SIT entries
+ */
+#define SIT_VBLOCK_MAP_SIZE 64
+#define SIT_ENTRY_PER_BLOCK (PAGE_CACHE_SIZE / sizeof(struct f2fs_sit_entry))
+
+struct f2fs_sit_entry {
+ __le16 vblocks;
+ __u8 valid_map[SIT_VBLOCK_MAP_SIZE];
+ __le64 mtime;
+} __packed;
+
+struct f2fs_sit_block {
+ struct f2fs_sit_entry entries[SIT_ENTRY_PER_BLOCK];
+} __packed;
+
+/**
+ * For segment summary
+ *
+ * NOTE : For initializing fields, you must use set_summary
+ *
+ * - If data page, nid represents dnode's nid
+ * - If node page, nid represents the node page's nid.
+ *
+ * The ofs_in_node is used by only data page. It represents offset
+ * from node's page's beginning to get a data block address.
+ * ex) data_blkaddr = (block_t)(nodepage_start_address + ofs_in_node)
+ */
+struct f2fs_summary {
+ __le32 nid;
+ union {
+ __u8 reserved[3];
+ struct {
+ __u8 version;
+ __le16 ofs_in_node;
+ } __packed;
+ };
+} __packed;
+
+struct summary_footer {
+ unsigned char entry_type;
+ __u32 check_sum;
+} __packed;
+
+#define SUMMARY_SIZE (sizeof(struct f2fs_summary))
+#define SUM_FOOTER_SIZE (sizeof(struct summary_footer))
+#define ENTRIES_IN_SUM 512
+#define SUM_ENTRY_SIZE (SUMMARY_SIZE * ENTRIES_IN_SUM)
+#define SUM_JOURNAL_SIZE (PAGE_CACHE_SIZE - SUM_FOOTER_SIZE -\
+ SUM_ENTRY_SIZE)
+struct nat_journal_entry {
+ __le32 nid;
+ struct f2fs_nat_entry ne;
+} __packed;
+
+struct sit_journal_entry {
+ __le32 segno;
+ struct f2fs_sit_entry se;
+} __packed;
+
+#define NAT_JOURNAL_ENTRIES ((SUM_JOURNAL_SIZE - 2) /\
+ sizeof(struct nat_journal_entry))
+#define NAT_JOURNAL_RESERVED ((SUM_JOURNAL_SIZE - 2) %\
+ sizeof(struct nat_journal_entry))
+#define SIT_JOURNAL_ENTRIES ((SUM_JOURNAL_SIZE - 2) /\
+ sizeof(struct sit_journal_entry))
+#define SIT_JOURNAL_RESERVED ((SUM_JOURNAL_SIZE - 2) %\
+ sizeof(struct sit_journal_entry))
+enum {
+ NAT_JOURNAL = 0,
+ SIT_JOURNAL
+};
+
+struct nat_journal {
+ struct nat_journal_entry entries[NAT_JOURNAL_ENTRIES];
+ __u8 reserved[NAT_JOURNAL_RESERVED];
+} __packed;
+
+struct sit_journal {
+ struct sit_journal_entry entries[SIT_JOURNAL_ENTRIES];
+ __u8 reserved[SIT_JOURNAL_RESERVED];
+} __packed;
+
+struct f2fs_summary_block {
+ struct f2fs_summary entries[ENTRIES_IN_SUM];
+ union {
+ __le16 n_nats;
+ __le16 n_sits;
+ };
+ union {
+ struct nat_journal nat_j;
+ struct sit_journal sit_j;
+ };
+ struct summary_footer footer;
+} __packed;
+
+/*
+ * For directory operations
+ */
+#define F2FS_DOT_HASH 0
+#define F2FS_DDOT_HASH F2FS_DOT_HASH
+#define F2FS_MAX_HASH (~((0x3ULL) << 62))
+#define F2FS_HASH_COL_BIT ((0x1ULL) << 63)
+
+typedef __le32 f2fs_hash_t;
+
+#define F2FS_NAME_LEN 8
+#define NR_DENTRY_IN_BLOCK 214 /* the number of dentry in a block */
+#define MAX_DIR_HASH_DEPTH 63 /* MAX level for dir lookup */
+
+#define SIZE_OF_DIR_ENTRY 11 /* by byte */
+#define SIZE_OF_DENTRY_BITMAP ((NR_DENTRY_IN_BLOCK + BITS_PER_BYTE - 1) / \
+ BITS_PER_BYTE)
+#define SIZE_OF_RESERVED (PAGE_SIZE - ((SIZE_OF_DIR_ENTRY + \
+ F2FS_NAME_LEN) * \
+ NR_DENTRY_IN_BLOCK + SIZE_OF_DENTRY_BITMAP))
+
+struct f2fs_dir_entry {
+ __le32 hash_code; /* hash code of file name */
+ __le32 ino; /* node number of inode */
+ __le16 name_len; /* the size of file name
+ length in unicode characters */
+ __u8 file_type;
+} __packed;
+
+struct f2fs_dentry_block {
+ __u8 dentry_bitmap[SIZE_OF_DENTRY_BITMAP];
+ __u8 reserved[SIZE_OF_RESERVED];
+ struct f2fs_dir_entry dentry[NR_DENTRY_IN_BLOCK];
+ __u8 filename[NR_DENTRY_IN_BLOCK][F2FS_NAME_LEN];
+} __packed;
+
+enum {
+ F2FS_FT_UNKNOWN,
+ F2FS_FT_REG_FILE,
+ F2FS_FT_DIR,
+ F2FS_FT_CHRDEV,
+ F2FS_FT_BLKDEV,
+ F2FS_FT_FIFO,
+ F2FS_FT_SOCK,
+ F2FS_FT_SYMLINK,
+ F2FS_FT_MAX
+};
+
+#endif /* _LINUX_F2FS_FS_H */
--
1.7.9.5
---
Jaegeuk Kim
Samsung
Am Freitag, 5. Oktober 2012 schrieb 김재극:
> This adds a header file describing the on-disk layout of f2fs.
>
> Signed-off-by: Changman Lee <[email protected]>
> Signed-off-by: Chul Lee <[email protected]>
> Signed-off-by: Jaegeuk Kim <[email protected]>
> ---
> include/linux/f2fs_fs.h | 359 +++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 359 insertions(+)
> create mode 100644 include/linux/f2fs_fs.h
>
> diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h
> new file mode 100644
> index 0000000..b17eeec
> --- /dev/null
> +++ b/include/linux/f2fs_fs.h
> @@ -0,0 +1,359 @@
> +/**
> + * include/linux/f2fs_fs.h
[…]
> +/*
> + * For superblock
> + */
> +struct f2fs_super_block {
> + __le32 magic; /* Magic Number */
> + __le16 major_ver; /* Major Version */
> + __le16 minor_ver; /* Minor Version */
> + __le32 log_sectorsize; /* log2 (Sector size in bytes) */
> + __le32 log_sectors_per_block; /* log2 (Number of sectors per block */
> + __le32 log_blocksize; /* log2 (Block size in bytes) */
> + __le32 log_blocks_per_seg; /* log2 (Number of blocks per segment) */
> + __le32 log_segs_per_sec; /* log2 (Number of segments per section) */
> + __le32 secs_per_zone; /* Number of sections per zone */
> + __le32 checksum_offset; /* Checksum position in this super block */
> + __le64 block_count; /* Total number of blocks */
> + __le32 section_count; /* Total number of sections */
> + __le32 segment_count; /* Total number of segments */
> + __le32 segment_count_ckpt; /* Total number of segments
> + in Checkpoint area */
> + __le32 segment_count_sit; /* Total number of segments
> + in Segment information table */
> + __le32 segment_count_nat; /* Total number of segments
> + in Node address table */
> + /*Total number of segments in Segment summary area */
> + __le32 segment_count_ssa;
> + /* Total number of segments in Main area */
> + __le32 segment_count_main;
> + __le32 failure_safe_block_distance;
> + __le64 segment0_blkaddr; /* Start block address of Segment 0 */
> + __le64 start_segment_checkpoint; /* Start block address of ckpt */
> + __le64 sit_blkaddr; /* Start block address of SIT */
> + __le64 nat_blkaddr; /* Start block address of NAT */
> + __le64 ssa_blkaddr; /* Start block address of SSA */
> + __le64 main_blkaddr; /* Start block address of Main area */
> + __le32 root_ino; /* Root directory inode number */
> + __le32 node_ino; /* node inode number */
> + __le32 meta_ino; /* meta inode number */
> + __le32 volume_serial_number; /* VSN is optional field */
> + __le16 volume_name[8]; /* Volume Name. 8 unicode characters */
> +} __packed;
Any reason why volume label is that short?
If superblock can occupy up to 4096 bytes, i.e. a block, I think a bit
more space for volume label would fit in there. I count 128 bytes up
to volume_name. Do other structures need to fit in superblock as well?
I believe users easily can hit a limit of 8 unicode characters even for
USB sticks and SD cards.
Nice to see a flash friendly filesystem which seems to take lots of the
stuff in account I read at Linaro website about flash devices.
--
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7
On Fri, 2012-10-05 at 20:56 +0900, 김재극 wrote:
> + __le32 i_atime; /* Access time */
> + __le32 i_ctime; /* inode Change time */
> + __le32 i_mtime; /* Modification time */
> + __le32 i_btime; /* file creation time*/
It's probably to late, but have you considered using 64 bit timestamps?
In LanyFS we use 64 bit and I'm very interested in reasoning for 32 bit
timestamps.
Regards
Dan
--
Dan Luedtke
http://www.danrl.de
On 10/05/2012 10:54 AM, Dan Luedtke wrote:
> On Fri, 2012-10-05 at 20:56 +0900, 김재극 wrote:
>> + __le32 i_atime; /* Access time */
>> + __le32 i_ctime; /* inode Change time */
>> + __le32 i_mtime; /* Modification time */
>> + __le32 i_btime; /* file creation time*/
>
> It's probably to late, but have you considered using 64 bit timestamps?
> In LanyFS we use 64 bit and I'm very interested in reasoning for 32 bit
> timestamps.
>
Yes and please don't make them millisecond resolution. This is a real headache.
>From the like of make to NFS to reboot/corruption recovery... the list is endless
Please make it nanoseconds resolution.
> Regards
>
> Dan
>
Thanks
Boaz
2012-10-05 (금), 14:09 -0700, Boaz Harrosh:
> On 10/05/2012 10:54 AM, Dan Luedtke wrote:
> > On Fri, 2012-10-05 at 20:56 +0900, 김재극 wrote:
> >> + __le32 i_atime; /* Access time */
> >> + __le32 i_ctime; /* inode Change time */
> >> + __le32 i_mtime; /* Modification time */
> >> + __le32 i_btime; /* file creation time*/
> >
> > It's probably to late, but have you considered using 64 bit timestamps?
> > In LanyFS we use 64 bit and I'm very interested in reasoning for 32 bit
> > timestamps.
> >
>
> Yes and please don't make them millisecond resolution. This is a real headache.
> From the like of make to NFS to reboot/corruption recovery... the list is endless
>
> Please make it nanoseconds resolution.
>
Ok, I'll take a look at this.
Thanks,
> > Regards
> >
> > Dan
> >
>
> Thanks
> Boaz
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
Jaegeuk Kim
Samsung
2012-10-05 (금), 19:46 +0200, Martin Steigerwald:
> Am Freitag, 5. Oktober 2012 schrieb 김재극:
> > This adds a header file describing the on-disk layout of f2fs.
> >
> > Signed-off-by: Changman Lee <[email protected]>
> > Signed-off-by: Chul Lee <[email protected]>
> > Signed-off-by: Jaegeuk Kim <[email protected]>
> > ---
> > include/linux/f2fs_fs.h | 359 +++++++++++++++++++++++++++++++++++++++++++++++
> > 1 file changed, 359 insertions(+)
> > create mode 100644 include/linux/f2fs_fs.h
> >
> > diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h
> > new file mode 100644
> > index 0000000..b17eeec
> > --- /dev/null
> > +++ b/include/linux/f2fs_fs.h
> > @@ -0,0 +1,359 @@
> > +/**
> > + * include/linux/f2fs_fs.h
> […]
> > +/*
> > + * For superblock
> > + */
> > +struct f2fs_super_block {
> > + __le32 magic; /* Magic Number */
> > + __le16 major_ver; /* Major Version */
> > + __le16 minor_ver; /* Minor Version */
> > + __le32 log_sectorsize; /* log2 (Sector size in bytes) */
> > + __le32 log_sectors_per_block; /* log2 (Number of sectors per block */
> > + __le32 log_blocksize; /* log2 (Block size in bytes) */
> > + __le32 log_blocks_per_seg; /* log2 (Number of blocks per segment) */
> > + __le32 log_segs_per_sec; /* log2 (Number of segments per section) */
> > + __le32 secs_per_zone; /* Number of sections per zone */
> > + __le32 checksum_offset; /* Checksum position in this super block */
> > + __le64 block_count; /* Total number of blocks */
> > + __le32 section_count; /* Total number of sections */
> > + __le32 segment_count; /* Total number of segments */
> > + __le32 segment_count_ckpt; /* Total number of segments
> > + in Checkpoint area */
> > + __le32 segment_count_sit; /* Total number of segments
> > + in Segment information table */
> > + __le32 segment_count_nat; /* Total number of segments
> > + in Node address table */
> > + /*Total number of segments in Segment summary area */
> > + __le32 segment_count_ssa;
> > + /* Total number of segments in Main area */
> > + __le32 segment_count_main;
> > + __le32 failure_safe_block_distance;
> > + __le64 segment0_blkaddr; /* Start block address of Segment 0 */
> > + __le64 start_segment_checkpoint; /* Start block address of ckpt */
> > + __le64 sit_blkaddr; /* Start block address of SIT */
> > + __le64 nat_blkaddr; /* Start block address of NAT */
> > + __le64 ssa_blkaddr; /* Start block address of SSA */
> > + __le64 main_blkaddr; /* Start block address of Main area */
> > + __le32 root_ino; /* Root directory inode number */
> > + __le32 node_ino; /* node inode number */
> > + __le32 meta_ino; /* meta inode number */
> > + __le32 volume_serial_number; /* VSN is optional field */
> > + __le16 volume_name[8]; /* Volume Name. 8 unicode characters */
> > +} __packed;
>
> Any reason why volume label is that short?
>
Noop.
> If superblock can occupy up to 4096 bytes, i.e. a block, I think a bit
> more space for volume label would fit in there. I count 128 bytes up
> to volume_name. Do other structures need to fit in superblock as well?
>
I think it cannot be a problem to give more bytes to the name space.
I'll do this.
> I believe users easily can hit a limit of 8 unicode characters even for
> USB sticks and SD cards.
> Nice to see a flash friendly filesystem which seems to take lots of the
> stuff in account I read at Linaro website about flash devices.
>
Thank you.
--
Jaegeuk Kim
Samsung
On Fri, Oct 05, 2012 at 08:56:44PM +0900, ????????? wrote:
> +struct node_footer {
> + __le32 nid; /* node id */
> + __le32 ino; /* inode nunmber */
> + __le32 cold:1; /* cold mark */
> + __le32 fsync:1; /* fsync mark */
> + __le32 dentry:1; /* dentry mark */
> + __le32 offset:29; /* offset in inode's node space */
A bitfield for a on-disk structure? This will have endianity issues,
(but I don't know if you intend to support big-endian). It's not enough
to use cpu_to_le* as in
fill_node_footer(...) {
rn->footer.offset = cpu_to_le32(ofs);
}
because the bitfield inside the structure will be already defined
reversed. The cpu_to_le macro will only convert value of 'ofs' but will
place it to different bits than it would on a little-endian arch.
There are macros to define bitfields in an endian-neutral way (or do it
by #ifdefs though it also involves duplicating the item names), or you
can alternatively use two structs fr disk-only and memory-only access,
the disk one stores __le32 with value combined of all and the in-memory
gets set up properly and will look like your current version of the
structure.
(More about not using bitfields http://yarchive.net/comp/linux/bitfields.html)
> + __le64 cp_ver; /* checkpoint version */
> + __le32 next_blkaddr; /* next node page block address */
> +} __packed;
> +
david
> +/*
> + * For superblock
> + */
> +struct f2fs_super_block {
> + __le32 magic; /* Magic Number */
> + __le16 major_ver; /* Major Version */
> + __le16 minor_ver; /* Minor Version */
> + __le32 log_sectorsize; /* log2 (Sector size in bytes) */
> + __le32 log_sectors_per_block; /* log2 (Number of sectors per block */
> + __le32 log_blocksize; /* log2 (Block size in bytes) */
Why store log_blocksize on disk when it can be calculated from
log_sectorsize and log_sectors_per_block? It may be handy to keep this
in the in-memory superblock but keeping it on-disk means you need a
consistency check and error code path when loading the superblock.
> +struct f2fs_inode {
> + __le16 i_mode; /* File mode */
> + __le16 i_reserved; /* Reserved */
> + __le32 i_uid; /* User ID */
> + __le32 i_gid; /* Group ID */
> + __le32 i_links; /* Links count */
> + __le64 i_size; /* File size in bytes */
> + __le64 i_blocks; /* File size in bytes */
File size in blocks
> +struct direct_node {
> + __le32 addr[ADDRS_PER_BLOCK]; /* aray of data block address */
s/aray/array/
> +} __packed;
> +
> +struct indirect_node {
> + __le32 nid[NIDS_PER_BLOCK]; /* aray of data block address */
s/aray/array/
Stefan
Dear David Sterba,
David Sterba wrote:
> On Fri, Oct 05, 2012 at 08:56:44PM +0900, ????????? wrote:
> > +struct node_footer {
> > + __le32 nid; /* node id */
> > + __le32 ino; /* inode nunmber */
> > + __le32 cold:1; /* cold mark */
> > + __le32 fsync:1; /* fsync mark */
> > + __le32 dentry:1; /* dentry mark */
> > + __le32 offset:29; /* offset in inode's node space */
>
> A bitfield for a on-disk structure? This will have endianity issues,
> (but I don't know if you intend to support big-endian). It's not enough
> to use cpu_to_le* as in
>
> fill_node_footer(...) {
>
> rn->footer.offset = cpu_to_le32(ofs);
>
> }
>
> because the bitfield inside the structure will be already defined
> reversed. The cpu_to_le macro will only convert value of 'ofs' but will
> place it to different bits than it would on a little-endian arch.
>
> There are macros to define bitfields in an endian-neutral way (or do it
> by #ifdefs though it also involves duplicating the item names), or you
> can alternatively use two structs fr disk-only and memory-only access,
> the disk one stores __le32 with value combined of all and the in-memory
> gets set up properly and will look like your current version of the
> structure.
>
> (More about not using bitfields
> http://yarchive.net/comp/linux/bitfields.html)
I appreciate your kind feedback, but we should've avoided using bitfields
for on-disk structure. As you suggested, we'll revise them. The latter
approach would be good.
>
>
> david
Chul
[snip]
> > +/*
> > + * For superblock
> > + */
> > +struct f2fs_super_block {
> > + __le32 magic; /* Magic Number */
> > + __le16 major_ver; /* Major Version */
> > + __le16 minor_ver; /* Minor Version */
> > + __le32 log_sectorsize; /* log2 (Sector size in bytes) */
> > + __le32 log_sectors_per_block; /* log2 (Number of sectors per block */
> > + __le32 log_blocksize; /* log2 (Block size in bytes) */
>
> Why store log_blocksize on disk when it can be calculated from
> log_sectorsize and log_sectors_per_block? It may be handy to keep this
> in the in-memory superblock but keeping it on-disk means you need a
> consistency check and error code path when loading the superblock.
>
Yes, I added this for sanity check of superblock.
> > +struct f2fs_inode {
> > + __le16 i_mode; /* File mode */
> > + __le16 i_reserved; /* Reserved */
> > + __le32 i_uid; /* User ID */
> > + __le32 i_gid; /* Group ID */
> > + __le32 i_links; /* Links count */
> > + __le64 i_size; /* File size in bytes */
> > + __le64 i_blocks; /* File size in bytes */
>
> File size in blocks
>
> > +struct direct_node {
> > + __le32 addr[ADDRS_PER_BLOCK]; /* aray of data block address */
>
> s/aray/array/
>
> > +} __packed;
> > +
> > +struct indirect_node {
> > + __le32 nid[NIDS_PER_BLOCK]; /* aray of data block address */
>
> s/aray/array/
I'll change them.
Thank you.
>
> Stefan
On Friday 05 October 2012 20:56:44 김재극 wrote:
> +struct f2fs_nat_entry {
> + __u8 version;
> + __le32 ino;
> + __le32 block_addr;
> +} __packed;
> +
> +struct f2fs_nat_block {
> + struct f2fs_nat_entry entries[NAT_ENTRY_PER_BLOCK];
> +} __packed;
Using "__packed" on structure is rather inefficient on CPU architectures
that cannot do aligned accesses. I would suggest you remove this
attribute everywhere you can. The f2fs_nat_entry is particularly
suboptimal because it is 9 bytes long, and I'm not sure if this
can be reasonably changed to a multiple of four.
In all other cases, you should try to lay out the structures
so that each member is naturally aligned and you don't need any __packed
attributes, in particular for those that are accessed a lot.
Arnd
2012-10-12 (금), 21:05 +0000, Arnd Bergmann:
> On Friday 05 October 2012 20:56:44 김재극 wrote:
> > +struct f2fs_nat_entry {
> > + __u8 version;
> > + __le32 ino;
> > + __le32 block_addr;
> > +} __packed;
> > +
> > +struct f2fs_nat_block {
> > + struct f2fs_nat_entry entries[NAT_ENTRY_PER_BLOCK];
> > +} __packed;
>
> Using "__packed" on structure is rather inefficient on CPU architectures
> that cannot do aligned accesses. I would suggest you remove this
> attribute everywhere you can. The f2fs_nat_entry is particularly
> suboptimal because it is 9 bytes long, and I'm not sure if this
> can be reasonably changed to a multiple of four.
>
> In all other cases, you should try to lay out the structures
> so that each member is naturally aligned and you don't need any __packed
> attributes, in particular for those that are accessed a lot.
>
Ok, I'll check all the in-memory data structures.
Thank you for comments.
> Arnd
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
Jaegeuk Kim
Samsung