fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
// SPDX-License-Identifier: GPL-2.0
|
|
|
|
|
/*
|
|
|
|
|
* Simple file system for zoned block devices exposing zones as files.
|
|
|
|
|
*
|
|
|
|
|
* Copyright (C) 2019 Western Digital Corporation or its affiliates.
|
|
|
|
|
*/
|
|
|
|
|
#include <linux/module.h>
|
2021-06-29 10:36:30 +08:00
|
|
|
#include <linux/pagemap.h>
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
#include <linux/magic.h>
|
|
|
|
|
#include <linux/iomap.h>
|
|
|
|
|
#include <linux/init.h>
|
|
|
|
|
#include <linux/slab.h>
|
|
|
|
|
#include <linux/blkdev.h>
|
|
|
|
|
#include <linux/statfs.h>
|
|
|
|
|
#include <linux/writeback.h>
|
|
|
|
|
#include <linux/quotaops.h>
|
|
|
|
|
#include <linux/seq_file.h>
|
|
|
|
|
#include <linux/parser.h>
|
|
|
|
|
#include <linux/uio.h>
|
|
|
|
|
#include <linux/mman.h>
|
|
|
|
|
#include <linux/sched/mm.h>
|
|
|
|
|
#include <linux/crc32.h>
|
2020-05-12 16:55:54 +08:00
|
|
|
#include <linux/task_io_accounting_ops.h>
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
#include "zonefs.h"
|
|
|
|
|
|
2021-01-27 04:21:15 +08:00
|
|
|
#define CREATE_TRACE_POINTS
|
|
|
|
|
#include "trace.h"
|
|
|
|
|
|
2022-04-12 17:54:39 +08:00
|
|
|
/*
|
|
|
|
|
* Manage the active zone count. Called with zi->i_truncate_mutex held.
|
|
|
|
|
*/
|
2022-11-25 08:39:33 +08:00
|
|
|
void zonefs_account_active(struct inode *inode)
|
2022-04-12 17:54:39 +08:00
|
|
|
{
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(inode->i_sb);
|
|
|
|
|
struct zonefs_inode_info *zi = ZONEFS_I(inode);
|
|
|
|
|
|
|
|
|
|
lockdep_assert_held(&zi->i_truncate_mutex);
|
|
|
|
|
|
|
|
|
|
if (zi->i_ztype != ZONEFS_ZTYPE_SEQ)
|
|
|
|
|
return;
|
|
|
|
|
|
2022-11-21 15:29:37 +08:00
|
|
|
/*
|
|
|
|
|
* For zones that transitioned to the offline or readonly condition,
|
|
|
|
|
* we only need to clear the active state.
|
|
|
|
|
*/
|
|
|
|
|
if (zi->i_flags & (ZONEFS_ZONE_OFFLINE | ZONEFS_ZONE_READONLY))
|
|
|
|
|
goto out;
|
|
|
|
|
|
2022-04-12 17:54:39 +08:00
|
|
|
/*
|
|
|
|
|
* If the zone is active, that is, if it is explicitly open or
|
|
|
|
|
* partially written, check if it was already accounted as active.
|
|
|
|
|
*/
|
|
|
|
|
if ((zi->i_flags & ZONEFS_ZONE_OPEN) ||
|
|
|
|
|
(zi->i_wpoffset > 0 && zi->i_wpoffset < zi->i_max_size)) {
|
|
|
|
|
if (!(zi->i_flags & ZONEFS_ZONE_ACTIVE)) {
|
|
|
|
|
zi->i_flags |= ZONEFS_ZONE_ACTIVE;
|
|
|
|
|
atomic_inc(&sbi->s_active_seq_files);
|
|
|
|
|
}
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
2022-11-21 15:29:37 +08:00
|
|
|
out:
|
2022-04-12 17:54:39 +08:00
|
|
|
/* The zone is not active. If it was, update the active count */
|
|
|
|
|
if (zi->i_flags & ZONEFS_ZONE_ACTIVE) {
|
|
|
|
|
zi->i_flags &= ~ZONEFS_ZONE_ACTIVE;
|
|
|
|
|
atomic_dec(&sbi->s_active_seq_files);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2022-11-25 08:39:33 +08:00
|
|
|
int zonefs_zone_mgmt(struct inode *inode, enum req_op op)
|
2020-09-11 16:56:48 +08:00
|
|
|
{
|
|
|
|
|
struct zonefs_inode_info *zi = ZONEFS_I(inode);
|
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
|
|
lockdep_assert_held(&zi->i_truncate_mutex);
|
|
|
|
|
|
2022-04-12 16:41:37 +08:00
|
|
|
/*
|
|
|
|
|
* With ZNS drives, closing an explicitly open zone that has not been
|
|
|
|
|
* written will change the zone state to "closed", that is, the zone
|
|
|
|
|
* will remain active. Since this can then cause failure of explicit
|
|
|
|
|
* open operation on other zones if the drive active zone resources
|
|
|
|
|
* are exceeded, make sure that the zone does not remain active by
|
|
|
|
|
* resetting it.
|
|
|
|
|
*/
|
|
|
|
|
if (op == REQ_OP_ZONE_CLOSE && !zi->i_wpoffset)
|
|
|
|
|
op = REQ_OP_ZONE_RESET;
|
|
|
|
|
|
2021-01-27 04:21:15 +08:00
|
|
|
trace_zonefs_zone_mgmt(inode, op);
|
2020-09-11 16:56:48 +08:00
|
|
|
ret = blkdev_zone_mgmt(inode->i_sb->s_bdev, op, zi->i_zsector,
|
|
|
|
|
zi->i_zone_size >> SECTOR_SHIFT, GFP_NOFS);
|
|
|
|
|
if (ret) {
|
|
|
|
|
zonefs_err(inode->i_sb,
|
|
|
|
|
"Zone management operation %s at %llu failed %d\n",
|
|
|
|
|
blk_op_str(op), zi->i_zsector, ret);
|
|
|
|
|
return ret;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
2022-11-25 08:39:33 +08:00
|
|
|
void zonefs_i_size_write(struct inode *inode, loff_t isize)
|
zonefs: open/close zone on file open/close
NVMe Zoned Namespace introduced the concept of active zones, which are
zones in the implicit open, explicit open or closed condition. Drives may
have a limit on the number of zones that can be simultaneously active.
This potential limitation translate into a risk for applications to see
write IO errors due to this limit if the zone of a file being written to is
not already active when a write request is issued.
To avoid these potential errors, the zone of a file can explicitly be made
active using an open zone command when the file is open for the first
time. If the zone open command succeeds, the application is then
guaranteed that write requests can be processed. This indirect management
of active zones relies on the maximum number of open zones of a drive,
which is always lower or equal to the maximum number of active zones.
On the first open of a sequential zone file, send a REQ_OP_ZONE_OPEN
command to the block device. Conversely, on the last release of a zone
file and send a REQ_OP_ZONE_CLOSE to the device if the zone is not full or
empty.
As truncating a zone file to 0 or max can deactivate a zone as well, we
need to serialize against truncates and also be careful not to close a
zone as the file may still be open for writing, e.g. the user called
ftruncate(). If the zone file is not open and a process does a truncate(),
then no close operation is needed.
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
2020-09-11 16:56:50 +08:00
|
|
|
{
|
|
|
|
|
struct zonefs_inode_info *zi = ZONEFS_I(inode);
|
|
|
|
|
|
|
|
|
|
i_size_write(inode, isize);
|
|
|
|
|
/*
|
|
|
|
|
* A full zone is no longer open/active and does not need
|
|
|
|
|
* explicit closing.
|
|
|
|
|
*/
|
2022-04-12 17:54:39 +08:00
|
|
|
if (isize >= zi->i_max_size) {
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(inode->i_sb);
|
|
|
|
|
|
|
|
|
|
if (zi->i_flags & ZONEFS_ZONE_ACTIVE)
|
|
|
|
|
atomic_dec(&sbi->s_active_seq_files);
|
|
|
|
|
zi->i_flags &= ~(ZONEFS_ZONE_OPEN | ZONEFS_ZONE_ACTIVE);
|
|
|
|
|
}
|
zonefs: open/close zone on file open/close
NVMe Zoned Namespace introduced the concept of active zones, which are
zones in the implicit open, explicit open or closed condition. Drives may
have a limit on the number of zones that can be simultaneously active.
This potential limitation translate into a risk for applications to see
write IO errors due to this limit if the zone of a file being written to is
not already active when a write request is issued.
To avoid these potential errors, the zone of a file can explicitly be made
active using an open zone command when the file is open for the first
time. If the zone open command succeeds, the application is then
guaranteed that write requests can be processed. This indirect management
of active zones relies on the maximum number of open zones of a drive,
which is always lower or equal to the maximum number of active zones.
On the first open of a sequential zone file, send a REQ_OP_ZONE_OPEN
command to the block device. Conversely, on the last release of a zone
file and send a REQ_OP_ZONE_CLOSE to the device if the zone is not full or
empty.
As truncating a zone file to 0 or max can deactivate a zone as well, we
need to serialize against truncates and also be careful not to close a
zone as the file may still be open for writing, e.g. the user called
ftruncate(). If the zone file is not open and a process does a truncate(),
then no close operation is needed.
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
2020-09-11 16:56:50 +08:00
|
|
|
}
|
|
|
|
|
|
2022-11-25 08:39:33 +08:00
|
|
|
void zonefs_update_stats(struct inode *inode, loff_t new_isize)
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
{
|
|
|
|
|
struct super_block *sb = inode->i_sb;
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
|
|
|
|
|
loff_t old_isize = i_size_read(inode);
|
|
|
|
|
loff_t nr_blocks;
|
|
|
|
|
|
|
|
|
|
if (new_isize == old_isize)
|
|
|
|
|
return;
|
|
|
|
|
|
|
|
|
|
spin_lock(&sbi->s_lock);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* This may be called for an update after an IO error.
|
|
|
|
|
* So beware of the values seen.
|
|
|
|
|
*/
|
|
|
|
|
if (new_isize < old_isize) {
|
|
|
|
|
nr_blocks = (old_isize - new_isize) >> sb->s_blocksize_bits;
|
|
|
|
|
if (sbi->s_used_blocks > nr_blocks)
|
|
|
|
|
sbi->s_used_blocks -= nr_blocks;
|
|
|
|
|
else
|
|
|
|
|
sbi->s_used_blocks = 0;
|
|
|
|
|
} else {
|
|
|
|
|
sbi->s_used_blocks +=
|
|
|
|
|
(new_isize - old_isize) >> sb->s_blocksize_bits;
|
|
|
|
|
if (sbi->s_used_blocks > sbi->s_blocks)
|
|
|
|
|
sbi->s_used_blocks = sbi->s_blocks;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
spin_unlock(&sbi->s_lock);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Check a zone condition and adjust its file inode access permissions for
|
|
|
|
|
* offline and readonly zones. Return the inode size corresponding to the
|
|
|
|
|
* amount of readable data in the zone.
|
|
|
|
|
*/
|
|
|
|
|
static loff_t zonefs_check_zone_condition(struct inode *inode,
|
2020-03-20 20:36:54 +08:00
|
|
|
struct blk_zone *zone, bool warn,
|
|
|
|
|
bool mount)
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
{
|
|
|
|
|
struct zonefs_inode_info *zi = ZONEFS_I(inode);
|
|
|
|
|
|
|
|
|
|
switch (zone->cond) {
|
|
|
|
|
case BLK_ZONE_COND_OFFLINE:
|
|
|
|
|
/*
|
|
|
|
|
* Dead zone: make the inode immutable, disable all accesses
|
|
|
|
|
* and set the file size to 0 (zone wp set to zone start).
|
|
|
|
|
*/
|
|
|
|
|
if (warn)
|
|
|
|
|
zonefs_warn(inode->i_sb, "inode %lu: offline zone\n",
|
|
|
|
|
inode->i_ino);
|
|
|
|
|
inode->i_flags |= S_IMMUTABLE;
|
|
|
|
|
inode->i_mode &= ~0777;
|
|
|
|
|
zone->wp = zone->start;
|
2022-11-21 15:29:37 +08:00
|
|
|
zi->i_flags |= ZONEFS_ZONE_OFFLINE;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
return 0;
|
|
|
|
|
case BLK_ZONE_COND_READONLY:
|
2020-03-20 20:36:54 +08:00
|
|
|
/*
|
|
|
|
|
* The write pointer of read-only zones is invalid. If such a
|
|
|
|
|
* zone is found during mount, the file size cannot be retrieved
|
|
|
|
|
* so we treat the zone as offline (mount == true case).
|
|
|
|
|
* Otherwise, keep the file size as it was when last updated
|
|
|
|
|
* so that the user can recover data. In both cases, writes are
|
|
|
|
|
* always disabled for the zone.
|
|
|
|
|
*/
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
if (warn)
|
|
|
|
|
zonefs_warn(inode->i_sb, "inode %lu: read-only zone\n",
|
|
|
|
|
inode->i_ino);
|
|
|
|
|
inode->i_flags |= S_IMMUTABLE;
|
2020-03-20 20:36:54 +08:00
|
|
|
if (mount) {
|
|
|
|
|
zone->cond = BLK_ZONE_COND_OFFLINE;
|
|
|
|
|
inode->i_mode &= ~0777;
|
|
|
|
|
zone->wp = zone->start;
|
2022-11-21 15:29:37 +08:00
|
|
|
zi->i_flags |= ZONEFS_ZONE_OFFLINE;
|
2020-03-20 20:36:54 +08:00
|
|
|
return 0;
|
|
|
|
|
}
|
2022-11-21 15:29:37 +08:00
|
|
|
zi->i_flags |= ZONEFS_ZONE_READONLY;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
inode->i_mode &= ~0222;
|
2020-03-20 20:36:54 +08:00
|
|
|
return i_size_read(inode);
|
2021-02-17 17:58:11 +08:00
|
|
|
case BLK_ZONE_COND_FULL:
|
|
|
|
|
/* The write pointer of full zones is invalid. */
|
|
|
|
|
return zi->i_max_size;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
default:
|
|
|
|
|
if (zi->i_ztype == ZONEFS_ZTYPE_CNV)
|
|
|
|
|
return zi->i_max_size;
|
|
|
|
|
return (zone->wp - zone->start) << SECTOR_SHIFT;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
struct zonefs_ioerr_data {
|
|
|
|
|
struct inode *inode;
|
|
|
|
|
bool write;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static int zonefs_io_error_cb(struct blk_zone *zone, unsigned int idx,
|
|
|
|
|
void *data)
|
|
|
|
|
{
|
|
|
|
|
struct zonefs_ioerr_data *err = data;
|
|
|
|
|
struct inode *inode = err->inode;
|
|
|
|
|
struct zonefs_inode_info *zi = ZONEFS_I(inode);
|
|
|
|
|
struct super_block *sb = inode->i_sb;
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
|
|
|
|
|
loff_t isize, data_size;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Check the zone condition: if the zone is not "bad" (offline or
|
|
|
|
|
* read-only), read errors are simply signaled to the IO issuer as long
|
|
|
|
|
* as there is no inconsistency between the inode size and the amount of
|
|
|
|
|
* data writen in the zone (data_size).
|
|
|
|
|
*/
|
2020-03-20 20:36:54 +08:00
|
|
|
data_size = zonefs_check_zone_condition(inode, zone, true, false);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
isize = i_size_read(inode);
|
|
|
|
|
if (zone->cond != BLK_ZONE_COND_OFFLINE &&
|
|
|
|
|
zone->cond != BLK_ZONE_COND_READONLY &&
|
|
|
|
|
!err->write && isize == data_size)
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* At this point, we detected either a bad zone or an inconsistency
|
|
|
|
|
* between the inode size and the amount of data written in the zone.
|
|
|
|
|
* For the latter case, the cause may be a write IO error or an external
|
|
|
|
|
* action on the device. Two error patterns exist:
|
|
|
|
|
* 1) The inode size is lower than the amount of data in the zone:
|
|
|
|
|
* a write operation partially failed and data was writen at the end
|
|
|
|
|
* of the file. This can happen in the case of a large direct IO
|
|
|
|
|
* needing several BIOs and/or write requests to be processed.
|
|
|
|
|
* 2) The inode size is larger than the amount of data in the zone:
|
|
|
|
|
* this can happen with a deferred write error with the use of the
|
|
|
|
|
* device side write cache after getting successful write IO
|
|
|
|
|
* completions. Other possibilities are (a) an external corruption,
|
|
|
|
|
* e.g. an application reset the zone directly, or (b) the device
|
|
|
|
|
* has a serious problem (e.g. firmware bug).
|
|
|
|
|
*
|
|
|
|
|
* In all cases, warn about inode size inconsistency and handle the
|
|
|
|
|
* IO error according to the zone condition and to the mount options.
|
|
|
|
|
*/
|
|
|
|
|
if (zi->i_ztype == ZONEFS_ZTYPE_SEQ && isize != data_size)
|
|
|
|
|
zonefs_warn(sb, "inode %lu: invalid size %lld (should be %lld)\n",
|
|
|
|
|
inode->i_ino, isize, data_size);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* First handle bad zones signaled by hardware. The mount options
|
|
|
|
|
* errors=zone-ro and errors=zone-offline result in changing the
|
|
|
|
|
* zone condition to read-only and offline respectively, as if the
|
|
|
|
|
* condition was signaled by the hardware.
|
|
|
|
|
*/
|
|
|
|
|
if (zone->cond == BLK_ZONE_COND_OFFLINE ||
|
|
|
|
|
sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_ZOL) {
|
|
|
|
|
zonefs_warn(sb, "inode %lu: read/write access disabled\n",
|
|
|
|
|
inode->i_ino);
|
|
|
|
|
if (zone->cond != BLK_ZONE_COND_OFFLINE) {
|
|
|
|
|
zone->cond = BLK_ZONE_COND_OFFLINE;
|
|
|
|
|
data_size = zonefs_check_zone_condition(inode, zone,
|
2020-03-20 20:36:54 +08:00
|
|
|
false, false);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
} else if (zone->cond == BLK_ZONE_COND_READONLY ||
|
|
|
|
|
sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_ZRO) {
|
|
|
|
|
zonefs_warn(sb, "inode %lu: write access disabled\n",
|
|
|
|
|
inode->i_ino);
|
|
|
|
|
if (zone->cond != BLK_ZONE_COND_READONLY) {
|
|
|
|
|
zone->cond = BLK_ZONE_COND_READONLY;
|
|
|
|
|
data_size = zonefs_check_zone_condition(inode, zone,
|
2020-03-20 20:36:54 +08:00
|
|
|
false, false);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
2023-01-06 16:43:06 +08:00
|
|
|
} else if (sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_RO &&
|
|
|
|
|
data_size > isize) {
|
|
|
|
|
/* Do not expose garbage data */
|
|
|
|
|
data_size = isize;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
|
zonefs: open/close zone on file open/close
NVMe Zoned Namespace introduced the concept of active zones, which are
zones in the implicit open, explicit open or closed condition. Drives may
have a limit on the number of zones that can be simultaneously active.
This potential limitation translate into a risk for applications to see
write IO errors due to this limit if the zone of a file being written to is
not already active when a write request is issued.
To avoid these potential errors, the zone of a file can explicitly be made
active using an open zone command when the file is open for the first
time. If the zone open command succeeds, the application is then
guaranteed that write requests can be processed. This indirect management
of active zones relies on the maximum number of open zones of a drive,
which is always lower or equal to the maximum number of active zones.
On the first open of a sequential zone file, send a REQ_OP_ZONE_OPEN
command to the block device. Conversely, on the last release of a zone
file and send a REQ_OP_ZONE_CLOSE to the device if the zone is not full or
empty.
As truncating a zone file to 0 or max can deactivate a zone as well, we
need to serialize against truncates and also be careful not to close a
zone as the file may still be open for writing, e.g. the user called
ftruncate(). If the zone file is not open and a process does a truncate(),
then no close operation is needed.
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
2020-09-11 16:56:50 +08:00
|
|
|
/*
|
|
|
|
|
* If the filesystem is mounted with the explicit-open mount option, we
|
|
|
|
|
* need to clear the ZONEFS_ZONE_OPEN flag if the zone transitioned to
|
|
|
|
|
* the read-only or offline condition, to avoid attempting an explicit
|
|
|
|
|
* close of the zone when the inode file is closed.
|
|
|
|
|
*/
|
|
|
|
|
if ((sbi->s_mount_opts & ZONEFS_MNTOPT_EXPLICIT_OPEN) &&
|
|
|
|
|
(zone->cond == BLK_ZONE_COND_OFFLINE ||
|
|
|
|
|
zone->cond == BLK_ZONE_COND_READONLY))
|
|
|
|
|
zi->i_flags &= ~ZONEFS_ZONE_OPEN;
|
|
|
|
|
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
/*
|
|
|
|
|
* If error=remount-ro was specified, any error result in remounting
|
|
|
|
|
* the volume as read-only.
|
|
|
|
|
*/
|
|
|
|
|
if ((sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_RO) && !sb_rdonly(sb)) {
|
|
|
|
|
zonefs_warn(sb, "remounting filesystem read-only\n");
|
|
|
|
|
sb->s_flags |= SB_RDONLY;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Update block usage stats and the inode size to prevent access to
|
|
|
|
|
* invalid data.
|
|
|
|
|
*/
|
|
|
|
|
zonefs_update_stats(inode, data_size);
|
zonefs: open/close zone on file open/close
NVMe Zoned Namespace introduced the concept of active zones, which are
zones in the implicit open, explicit open or closed condition. Drives may
have a limit on the number of zones that can be simultaneously active.
This potential limitation translate into a risk for applications to see
write IO errors due to this limit if the zone of a file being written to is
not already active when a write request is issued.
To avoid these potential errors, the zone of a file can explicitly be made
active using an open zone command when the file is open for the first
time. If the zone open command succeeds, the application is then
guaranteed that write requests can be processed. This indirect management
of active zones relies on the maximum number of open zones of a drive,
which is always lower or equal to the maximum number of active zones.
On the first open of a sequential zone file, send a REQ_OP_ZONE_OPEN
command to the block device. Conversely, on the last release of a zone
file and send a REQ_OP_ZONE_CLOSE to the device if the zone is not full or
empty.
As truncating a zone file to 0 or max can deactivate a zone as well, we
need to serialize against truncates and also be careful not to close a
zone as the file may still be open for writing, e.g. the user called
ftruncate(). If the zone file is not open and a process does a truncate(),
then no close operation is needed.
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
2020-09-11 16:56:50 +08:00
|
|
|
zonefs_i_size_write(inode, data_size);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
zi->i_wpoffset = data_size;
|
2022-04-12 17:54:39 +08:00
|
|
|
zonefs_account_active(inode);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* When an file IO error occurs, check the file zone to see if there is a change
|
|
|
|
|
* in the zone condition (e.g. offline or read-only). For a failed write to a
|
|
|
|
|
* sequential zone, the zone write pointer position must also be checked to
|
|
|
|
|
* eventually correct the file size and zonefs inode write pointer offset
|
|
|
|
|
* (which can be out of sync with the drive due to partial write failures).
|
|
|
|
|
*/
|
2022-11-25 08:39:33 +08:00
|
|
|
void __zonefs_io_error(struct inode *inode, bool write)
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
{
|
|
|
|
|
struct zonefs_inode_info *zi = ZONEFS_I(inode);
|
|
|
|
|
struct super_block *sb = inode->i_sb;
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
|
|
|
|
|
unsigned int noio_flag;
|
2022-10-25 12:39:31 +08:00
|
|
|
unsigned int nr_zones = 1;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
struct zonefs_ioerr_data err = {
|
|
|
|
|
.inode = inode,
|
|
|
|
|
.write = write,
|
|
|
|
|
};
|
|
|
|
|
int ret;
|
|
|
|
|
|
2022-10-25 12:39:31 +08:00
|
|
|
/*
|
|
|
|
|
* The only files that have more than one zone are conventional zone
|
|
|
|
|
* files with aggregated conventional zones, for which the inode zone
|
|
|
|
|
* size is always larger than the device zone size.
|
|
|
|
|
*/
|
|
|
|
|
if (zi->i_zone_size > bdev_zone_sectors(sb->s_bdev))
|
|
|
|
|
nr_zones = zi->i_zone_size >>
|
|
|
|
|
(sbi->s_zone_sectors_shift + SECTOR_SHIFT);
|
|
|
|
|
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
/*
|
|
|
|
|
* Memory allocations in blkdev_report_zones() can trigger a memory
|
|
|
|
|
* reclaim which may in turn cause a recursion into zonefs as well as
|
|
|
|
|
* struct request allocations for the same device. The former case may
|
|
|
|
|
* end up in a deadlock on the inode truncate mutex, while the latter
|
|
|
|
|
* may prevent IO forward progress. Executing the report zones under
|
|
|
|
|
* the GFP_NOIO context avoids both problems.
|
|
|
|
|
*/
|
|
|
|
|
noio_flag = memalloc_noio_save();
|
|
|
|
|
ret = blkdev_report_zones(sb->s_bdev, zi->i_zsector, nr_zones,
|
|
|
|
|
zonefs_io_error_cb, &err);
|
|
|
|
|
if (ret != nr_zones)
|
|
|
|
|
zonefs_err(sb, "Get inode %lu zone information failed %d\n",
|
|
|
|
|
inode->i_ino, ret);
|
|
|
|
|
memalloc_noio_restore(noio_flag);
|
2020-09-11 16:56:49 +08:00
|
|
|
}
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
static struct kmem_cache *zonefs_inode_cachep;
|
|
|
|
|
|
|
|
|
|
static struct inode *zonefs_alloc_inode(struct super_block *sb)
|
|
|
|
|
{
|
|
|
|
|
struct zonefs_inode_info *zi;
|
|
|
|
|
|
2022-03-23 05:41:03 +08:00
|
|
|
zi = alloc_inode_sb(sb, zonefs_inode_cachep, GFP_KERNEL);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
if (!zi)
|
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
|
|
inode_init_once(&zi->i_vnode);
|
|
|
|
|
mutex_init(&zi->i_truncate_mutex);
|
zonefs: open/close zone on file open/close
NVMe Zoned Namespace introduced the concept of active zones, which are
zones in the implicit open, explicit open or closed condition. Drives may
have a limit on the number of zones that can be simultaneously active.
This potential limitation translate into a risk for applications to see
write IO errors due to this limit if the zone of a file being written to is
not already active when a write request is issued.
To avoid these potential errors, the zone of a file can explicitly be made
active using an open zone command when the file is open for the first
time. If the zone open command succeeds, the application is then
guaranteed that write requests can be processed. This indirect management
of active zones relies on the maximum number of open zones of a drive,
which is always lower or equal to the maximum number of active zones.
On the first open of a sequential zone file, send a REQ_OP_ZONE_OPEN
command to the block device. Conversely, on the last release of a zone
file and send a REQ_OP_ZONE_CLOSE to the device if the zone is not full or
empty.
As truncating a zone file to 0 or max can deactivate a zone as well, we
need to serialize against truncates and also be careful not to close a
zone as the file may still be open for writing, e.g. the user called
ftruncate(). If the zone file is not open and a process does a truncate(),
then no close operation is needed.
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
2020-09-11 16:56:50 +08:00
|
|
|
zi->i_wr_refcnt = 0;
|
2022-04-12 19:52:35 +08:00
|
|
|
zi->i_flags = 0;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
return &zi->i_vnode;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void zonefs_free_inode(struct inode *inode)
|
|
|
|
|
{
|
|
|
|
|
kmem_cache_free(zonefs_inode_cachep, ZONEFS_I(inode));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* File system stat.
|
|
|
|
|
*/
|
|
|
|
|
static int zonefs_statfs(struct dentry *dentry, struct kstatfs *buf)
|
|
|
|
|
{
|
|
|
|
|
struct super_block *sb = dentry->d_sb;
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
|
|
|
|
|
enum zonefs_ztype t;
|
|
|
|
|
|
|
|
|
|
buf->f_type = ZONEFS_MAGIC;
|
|
|
|
|
buf->f_bsize = sb->s_blocksize;
|
|
|
|
|
buf->f_namelen = ZONEFS_NAME_MAX;
|
|
|
|
|
|
|
|
|
|
spin_lock(&sbi->s_lock);
|
|
|
|
|
|
|
|
|
|
buf->f_blocks = sbi->s_blocks;
|
|
|
|
|
if (WARN_ON(sbi->s_used_blocks > sbi->s_blocks))
|
|
|
|
|
buf->f_bfree = 0;
|
|
|
|
|
else
|
|
|
|
|
buf->f_bfree = buf->f_blocks - sbi->s_used_blocks;
|
|
|
|
|
buf->f_bavail = buf->f_bfree;
|
|
|
|
|
|
|
|
|
|
for (t = 0; t < ZONEFS_ZTYPE_MAX; t++) {
|
|
|
|
|
if (sbi->s_nr_files[t])
|
|
|
|
|
buf->f_files += sbi->s_nr_files[t] + 1;
|
|
|
|
|
}
|
|
|
|
|
buf->f_ffree = 0;
|
|
|
|
|
|
|
|
|
|
spin_unlock(&sbi->s_lock);
|
|
|
|
|
|
2021-03-23 01:39:43 +08:00
|
|
|
buf->f_fsid = uuid_to_fsid(sbi->s_uuid.b);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
enum {
|
|
|
|
|
Opt_errors_ro, Opt_errors_zro, Opt_errors_zol, Opt_errors_repair,
|
zonefs: open/close zone on file open/close
NVMe Zoned Namespace introduced the concept of active zones, which are
zones in the implicit open, explicit open or closed condition. Drives may
have a limit on the number of zones that can be simultaneously active.
This potential limitation translate into a risk for applications to see
write IO errors due to this limit if the zone of a file being written to is
not already active when a write request is issued.
To avoid these potential errors, the zone of a file can explicitly be made
active using an open zone command when the file is open for the first
time. If the zone open command succeeds, the application is then
guaranteed that write requests can be processed. This indirect management
of active zones relies on the maximum number of open zones of a drive,
which is always lower or equal to the maximum number of active zones.
On the first open of a sequential zone file, send a REQ_OP_ZONE_OPEN
command to the block device. Conversely, on the last release of a zone
file and send a REQ_OP_ZONE_CLOSE to the device if the zone is not full or
empty.
As truncating a zone file to 0 or max can deactivate a zone as well, we
need to serialize against truncates and also be careful not to close a
zone as the file may still be open for writing, e.g. the user called
ftruncate(). If the zone file is not open and a process does a truncate(),
then no close operation is needed.
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
2020-09-11 16:56:50 +08:00
|
|
|
Opt_explicit_open, Opt_err,
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static const match_table_t tokens = {
|
|
|
|
|
{ Opt_errors_ro, "errors=remount-ro"},
|
|
|
|
|
{ Opt_errors_zro, "errors=zone-ro"},
|
|
|
|
|
{ Opt_errors_zol, "errors=zone-offline"},
|
|
|
|
|
{ Opt_errors_repair, "errors=repair"},
|
zonefs: open/close zone on file open/close
NVMe Zoned Namespace introduced the concept of active zones, which are
zones in the implicit open, explicit open or closed condition. Drives may
have a limit on the number of zones that can be simultaneously active.
This potential limitation translate into a risk for applications to see
write IO errors due to this limit if the zone of a file being written to is
not already active when a write request is issued.
To avoid these potential errors, the zone of a file can explicitly be made
active using an open zone command when the file is open for the first
time. If the zone open command succeeds, the application is then
guaranteed that write requests can be processed. This indirect management
of active zones relies on the maximum number of open zones of a drive,
which is always lower or equal to the maximum number of active zones.
On the first open of a sequential zone file, send a REQ_OP_ZONE_OPEN
command to the block device. Conversely, on the last release of a zone
file and send a REQ_OP_ZONE_CLOSE to the device if the zone is not full or
empty.
As truncating a zone file to 0 or max can deactivate a zone as well, we
need to serialize against truncates and also be careful not to close a
zone as the file may still be open for writing, e.g. the user called
ftruncate(). If the zone file is not open and a process does a truncate(),
then no close operation is needed.
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
2020-09-11 16:56:50 +08:00
|
|
|
{ Opt_explicit_open, "explicit-open" },
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
{ Opt_err, NULL}
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static int zonefs_parse_options(struct super_block *sb, char *options)
|
|
|
|
|
{
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
|
|
|
|
|
substring_t args[MAX_OPT_ARGS];
|
|
|
|
|
char *p;
|
|
|
|
|
|
|
|
|
|
if (!options)
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
|
|
while ((p = strsep(&options, ",")) != NULL) {
|
|
|
|
|
int token;
|
|
|
|
|
|
|
|
|
|
if (!*p)
|
|
|
|
|
continue;
|
|
|
|
|
|
|
|
|
|
token = match_token(p, tokens, args);
|
|
|
|
|
switch (token) {
|
|
|
|
|
case Opt_errors_ro:
|
|
|
|
|
sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
|
|
|
|
|
sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_RO;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_errors_zro:
|
|
|
|
|
sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
|
|
|
|
|
sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_ZRO;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_errors_zol:
|
|
|
|
|
sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
|
|
|
|
|
sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_ZOL;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_errors_repair:
|
|
|
|
|
sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
|
|
|
|
|
sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_REPAIR;
|
|
|
|
|
break;
|
zonefs: open/close zone on file open/close
NVMe Zoned Namespace introduced the concept of active zones, which are
zones in the implicit open, explicit open or closed condition. Drives may
have a limit on the number of zones that can be simultaneously active.
This potential limitation translate into a risk for applications to see
write IO errors due to this limit if the zone of a file being written to is
not already active when a write request is issued.
To avoid these potential errors, the zone of a file can explicitly be made
active using an open zone command when the file is open for the first
time. If the zone open command succeeds, the application is then
guaranteed that write requests can be processed. This indirect management
of active zones relies on the maximum number of open zones of a drive,
which is always lower or equal to the maximum number of active zones.
On the first open of a sequential zone file, send a REQ_OP_ZONE_OPEN
command to the block device. Conversely, on the last release of a zone
file and send a REQ_OP_ZONE_CLOSE to the device if the zone is not full or
empty.
As truncating a zone file to 0 or max can deactivate a zone as well, we
need to serialize against truncates and also be careful not to close a
zone as the file may still be open for writing, e.g. the user called
ftruncate(). If the zone file is not open and a process does a truncate(),
then no close operation is needed.
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
2020-09-11 16:56:50 +08:00
|
|
|
case Opt_explicit_open:
|
|
|
|
|
sbi->s_mount_opts |= ZONEFS_MNTOPT_EXPLICIT_OPEN;
|
|
|
|
|
break;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
default:
|
|
|
|
|
return -EINVAL;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int zonefs_show_options(struct seq_file *seq, struct dentry *root)
|
|
|
|
|
{
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(root->d_sb);
|
|
|
|
|
|
|
|
|
|
if (sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_RO)
|
|
|
|
|
seq_puts(seq, ",errors=remount-ro");
|
|
|
|
|
if (sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_ZRO)
|
|
|
|
|
seq_puts(seq, ",errors=zone-ro");
|
|
|
|
|
if (sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_ZOL)
|
|
|
|
|
seq_puts(seq, ",errors=zone-offline");
|
|
|
|
|
if (sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_REPAIR)
|
|
|
|
|
seq_puts(seq, ",errors=repair");
|
|
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int zonefs_remount(struct super_block *sb, int *flags, char *data)
|
|
|
|
|
{
|
|
|
|
|
sync_filesystem(sb);
|
|
|
|
|
|
|
|
|
|
return zonefs_parse_options(sb, data);
|
|
|
|
|
}
|
|
|
|
|
|
2022-11-25 08:39:33 +08:00
|
|
|
static int zonefs_inode_setattr(struct user_namespace *mnt_userns,
|
|
|
|
|
struct dentry *dentry, struct iattr *iattr)
|
|
|
|
|
{
|
|
|
|
|
struct inode *inode = d_inode(dentry);
|
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
|
|
if (unlikely(IS_IMMUTABLE(inode)))
|
|
|
|
|
return -EPERM;
|
|
|
|
|
|
|
|
|
|
ret = setattr_prepare(&init_user_ns, dentry, iattr);
|
|
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Since files and directories cannot be created nor deleted, do not
|
|
|
|
|
* allow setting any write attributes on the sub-directories grouping
|
|
|
|
|
* files by zone type.
|
|
|
|
|
*/
|
|
|
|
|
if ((iattr->ia_valid & ATTR_MODE) && S_ISDIR(inode->i_mode) &&
|
|
|
|
|
(iattr->ia_mode & 0222))
|
|
|
|
|
return -EPERM;
|
|
|
|
|
|
|
|
|
|
if (((iattr->ia_valid & ATTR_UID) &&
|
|
|
|
|
!uid_eq(iattr->ia_uid, inode->i_uid)) ||
|
|
|
|
|
((iattr->ia_valid & ATTR_GID) &&
|
|
|
|
|
!gid_eq(iattr->ia_gid, inode->i_gid))) {
|
|
|
|
|
ret = dquot_transfer(mnt_userns, inode, iattr);
|
|
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (iattr->ia_valid & ATTR_SIZE) {
|
|
|
|
|
ret = zonefs_file_truncate(inode, iattr->ia_size);
|
|
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
setattr_copy(&init_user_ns, inode, iattr);
|
|
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
static const struct inode_operations zonefs_dir_inode_operations = {
|
|
|
|
|
.lookup = simple_lookup,
|
|
|
|
|
.setattr = zonefs_inode_setattr,
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static void zonefs_init_dir_inode(struct inode *parent, struct inode *inode,
|
|
|
|
|
enum zonefs_ztype type)
|
|
|
|
|
{
|
|
|
|
|
struct super_block *sb = parent->i_sb;
|
|
|
|
|
|
2022-07-06 15:03:45 +08:00
|
|
|
inode->i_ino = bdev_nr_zones(sb->s_bdev) + type + 1;
|
2021-01-21 21:19:25 +08:00
|
|
|
inode_init_owner(&init_user_ns, inode, parent, S_IFDIR | 0555);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
inode->i_op = &zonefs_dir_inode_operations;
|
|
|
|
|
inode->i_fop = &simple_dir_operations;
|
|
|
|
|
set_nlink(inode, 2);
|
|
|
|
|
inc_nlink(parent);
|
|
|
|
|
}
|
|
|
|
|
|
2022-11-25 08:39:33 +08:00
|
|
|
static const struct inode_operations zonefs_file_inode_operations = {
|
|
|
|
|
.setattr = zonefs_inode_setattr,
|
|
|
|
|
};
|
|
|
|
|
|
2022-04-12 16:41:37 +08:00
|
|
|
static int zonefs_init_file_inode(struct inode *inode, struct blk_zone *zone,
|
|
|
|
|
enum zonefs_ztype type)
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
{
|
|
|
|
|
struct super_block *sb = inode->i_sb;
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
|
|
|
|
|
struct zonefs_inode_info *zi = ZONEFS_I(inode);
|
2022-05-24 07:29:39 +08:00
|
|
|
int ret = 0;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
inode->i_ino = zone->start >> sbi->s_zone_sectors_shift;
|
|
|
|
|
inode->i_mode = S_IFREG | sbi->s_perm;
|
|
|
|
|
|
|
|
|
|
zi->i_ztype = type;
|
|
|
|
|
zi->i_zsector = zone->start;
|
2020-07-21 20:10:26 +08:00
|
|
|
zi->i_zone_size = zone->len << SECTOR_SHIFT;
|
2022-10-25 12:39:31 +08:00
|
|
|
if (zi->i_zone_size > bdev_zone_sectors(sb->s_bdev) << SECTOR_SHIFT &&
|
|
|
|
|
!(sbi->s_features & ZONEFS_F_AGGRCNV)) {
|
|
|
|
|
zonefs_err(sb,
|
|
|
|
|
"zone size %llu doesn't match device's zone sectors %llu\n",
|
|
|
|
|
zi->i_zone_size,
|
|
|
|
|
bdev_zone_sectors(sb->s_bdev) << SECTOR_SHIFT);
|
|
|
|
|
return -EINVAL;
|
|
|
|
|
}
|
2020-07-21 20:10:26 +08:00
|
|
|
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
zi->i_max_size = min_t(loff_t, MAX_LFS_FILESIZE,
|
2020-07-21 20:10:26 +08:00
|
|
|
zone->capacity << SECTOR_SHIFT);
|
2020-03-20 20:36:54 +08:00
|
|
|
zi->i_wpoffset = zonefs_check_zone_condition(inode, zone, true, true);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
inode->i_uid = sbi->s_uid;
|
|
|
|
|
inode->i_gid = sbi->s_gid;
|
|
|
|
|
inode->i_size = zi->i_wpoffset;
|
2020-07-21 20:10:26 +08:00
|
|
|
inode->i_blocks = zi->i_max_size >> SECTOR_SHIFT;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
inode->i_op = &zonefs_file_inode_operations;
|
|
|
|
|
inode->i_fop = &zonefs_file_operations;
|
|
|
|
|
inode->i_mapping->a_ops = &zonefs_file_aops;
|
|
|
|
|
|
|
|
|
|
sb->s_maxbytes = max(zi->i_max_size, sb->s_maxbytes);
|
|
|
|
|
sbi->s_blocks += zi->i_max_size >> sb->s_blocksize_bits;
|
|
|
|
|
sbi->s_used_blocks += zi->i_wpoffset >> sb->s_blocksize_bits;
|
2022-04-12 16:41:37 +08:00
|
|
|
|
2022-04-12 17:54:39 +08:00
|
|
|
mutex_lock(&zi->i_truncate_mutex);
|
|
|
|
|
|
2022-04-12 16:41:37 +08:00
|
|
|
/*
|
|
|
|
|
* For sequential zones, make sure that any open zone is closed first
|
|
|
|
|
* to ensure that the initial number of open zones is 0, in sync with
|
|
|
|
|
* the open zone accounting done when the mount option
|
|
|
|
|
* ZONEFS_MNTOPT_EXPLICIT_OPEN is used.
|
|
|
|
|
*/
|
|
|
|
|
if (type == ZONEFS_ZTYPE_SEQ &&
|
|
|
|
|
(zone->cond == BLK_ZONE_COND_IMP_OPEN ||
|
|
|
|
|
zone->cond == BLK_ZONE_COND_EXP_OPEN)) {
|
|
|
|
|
ret = zonefs_zone_mgmt(inode, REQ_OP_ZONE_CLOSE);
|
2022-04-12 17:54:39 +08:00
|
|
|
if (ret)
|
|
|
|
|
goto unlock;
|
2022-04-12 16:41:37 +08:00
|
|
|
}
|
|
|
|
|
|
2022-04-12 17:54:39 +08:00
|
|
|
zonefs_account_active(inode);
|
|
|
|
|
|
|
|
|
|
unlock:
|
|
|
|
|
mutex_unlock(&zi->i_truncate_mutex);
|
|
|
|
|
|
2022-05-24 07:29:39 +08:00
|
|
|
return ret;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static struct dentry *zonefs_create_inode(struct dentry *parent,
|
|
|
|
|
const char *name, struct blk_zone *zone,
|
|
|
|
|
enum zonefs_ztype type)
|
|
|
|
|
{
|
|
|
|
|
struct inode *dir = d_inode(parent);
|
|
|
|
|
struct dentry *dentry;
|
|
|
|
|
struct inode *inode;
|
2022-10-25 12:39:31 +08:00
|
|
|
int ret = -ENOMEM;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
dentry = d_alloc_name(parent, name);
|
|
|
|
|
if (!dentry)
|
2022-10-25 12:39:31 +08:00
|
|
|
return ERR_PTR(ret);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
inode = new_inode(parent->d_sb);
|
|
|
|
|
if (!inode)
|
|
|
|
|
goto dput;
|
|
|
|
|
|
|
|
|
|
inode->i_ctime = inode->i_mtime = inode->i_atime = dir->i_ctime;
|
2022-04-12 16:41:37 +08:00
|
|
|
if (zone) {
|
|
|
|
|
ret = zonefs_init_file_inode(inode, zone, type);
|
|
|
|
|
if (ret) {
|
|
|
|
|
iput(inode);
|
|
|
|
|
goto dput;
|
|
|
|
|
}
|
|
|
|
|
} else {
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
zonefs_init_dir_inode(dir, inode, type);
|
2022-04-12 16:41:37 +08:00
|
|
|
}
|
|
|
|
|
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
d_add(dentry, inode);
|
|
|
|
|
dir->i_size++;
|
|
|
|
|
|
|
|
|
|
return dentry;
|
|
|
|
|
|
|
|
|
|
dput:
|
|
|
|
|
dput(dentry);
|
|
|
|
|
|
2022-10-25 12:39:31 +08:00
|
|
|
return ERR_PTR(ret);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
struct zonefs_zone_data {
|
|
|
|
|
struct super_block *sb;
|
|
|
|
|
unsigned int nr_zones[ZONEFS_ZTYPE_MAX];
|
|
|
|
|
struct blk_zone *zones;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Create a zone group and populate it with zone files.
|
|
|
|
|
*/
|
|
|
|
|
static int zonefs_create_zgroup(struct zonefs_zone_data *zd,
|
|
|
|
|
enum zonefs_ztype type)
|
|
|
|
|
{
|
|
|
|
|
struct super_block *sb = zd->sb;
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
|
|
|
|
|
struct blk_zone *zone, *next, *end;
|
|
|
|
|
const char *zgroup_name;
|
|
|
|
|
char *file_name;
|
2022-10-25 12:39:31 +08:00
|
|
|
struct dentry *dir, *dent;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
unsigned int n = 0;
|
2020-07-20 16:52:52 +08:00
|
|
|
int ret;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
/* If the group is empty, there is nothing to do */
|
|
|
|
|
if (!zd->nr_zones[type])
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
|
|
file_name = kmalloc(ZONEFS_NAME_MAX, GFP_KERNEL);
|
|
|
|
|
if (!file_name)
|
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
|
|
if (type == ZONEFS_ZTYPE_CNV)
|
|
|
|
|
zgroup_name = "cnv";
|
|
|
|
|
else
|
|
|
|
|
zgroup_name = "seq";
|
|
|
|
|
|
|
|
|
|
dir = zonefs_create_inode(sb->s_root, zgroup_name, NULL, type);
|
2022-10-25 12:39:31 +08:00
|
|
|
if (IS_ERR(dir)) {
|
|
|
|
|
ret = PTR_ERR(dir);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
goto free;
|
2020-07-20 16:52:52 +08:00
|
|
|
}
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* The first zone contains the super block: skip it.
|
|
|
|
|
*/
|
2022-07-06 15:03:45 +08:00
|
|
|
end = zd->zones + bdev_nr_zones(sb->s_bdev);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
for (zone = &zd->zones[1]; zone < end; zone = next) {
|
|
|
|
|
|
|
|
|
|
next = zone + 1;
|
|
|
|
|
if (zonefs_zone_type(zone) != type)
|
|
|
|
|
continue;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* For conventional zones, contiguous zones can be aggregated
|
|
|
|
|
* together to form larger files. Note that this overwrites the
|
|
|
|
|
* length of the first zone of the set of contiguous zones
|
|
|
|
|
* aggregated together. If one offline or read-only zone is
|
|
|
|
|
* found, assume that all zones aggregated have the same
|
|
|
|
|
* condition.
|
|
|
|
|
*/
|
|
|
|
|
if (type == ZONEFS_ZTYPE_CNV &&
|
|
|
|
|
(sbi->s_features & ZONEFS_F_AGGRCNV)) {
|
|
|
|
|
for (; next < end; next++) {
|
|
|
|
|
if (zonefs_zone_type(next) != type)
|
|
|
|
|
break;
|
|
|
|
|
zone->len += next->len;
|
2020-07-21 20:10:26 +08:00
|
|
|
zone->capacity += next->capacity;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
if (next->cond == BLK_ZONE_COND_READONLY &&
|
|
|
|
|
zone->cond != BLK_ZONE_COND_OFFLINE)
|
|
|
|
|
zone->cond = BLK_ZONE_COND_READONLY;
|
|
|
|
|
else if (next->cond == BLK_ZONE_COND_OFFLINE)
|
|
|
|
|
zone->cond = BLK_ZONE_COND_OFFLINE;
|
|
|
|
|
}
|
2020-07-21 20:10:26 +08:00
|
|
|
if (zone->capacity != zone->len) {
|
|
|
|
|
zonefs_err(sb, "Invalid conventional zone capacity\n");
|
|
|
|
|
ret = -EINVAL;
|
|
|
|
|
goto free;
|
|
|
|
|
}
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Use the file number within its group as file name.
|
|
|
|
|
*/
|
|
|
|
|
snprintf(file_name, ZONEFS_NAME_MAX - 1, "%u", n);
|
2022-10-25 12:39:31 +08:00
|
|
|
dent = zonefs_create_inode(dir, file_name, zone, type);
|
|
|
|
|
if (IS_ERR(dent)) {
|
|
|
|
|
ret = PTR_ERR(dent);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
goto free;
|
2020-07-20 16:52:52 +08:00
|
|
|
}
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
n++;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
zonefs_info(sb, "Zone group \"%s\" has %u file%s\n",
|
|
|
|
|
zgroup_name, n, n > 1 ? "s" : "");
|
|
|
|
|
|
|
|
|
|
sbi->s_nr_files[type] = n;
|
|
|
|
|
ret = 0;
|
|
|
|
|
|
|
|
|
|
free:
|
|
|
|
|
kfree(file_name);
|
|
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int zonefs_get_zone_info_cb(struct blk_zone *zone, unsigned int idx,
|
|
|
|
|
void *data)
|
|
|
|
|
{
|
|
|
|
|
struct zonefs_zone_data *zd = data;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Count the number of usable zones: the first zone at index 0 contains
|
|
|
|
|
* the super block and is ignored.
|
|
|
|
|
*/
|
|
|
|
|
switch (zone->type) {
|
|
|
|
|
case BLK_ZONE_TYPE_CONVENTIONAL:
|
|
|
|
|
zone->wp = zone->start + zone->len;
|
|
|
|
|
if (idx)
|
|
|
|
|
zd->nr_zones[ZONEFS_ZTYPE_CNV]++;
|
|
|
|
|
break;
|
|
|
|
|
case BLK_ZONE_TYPE_SEQWRITE_REQ:
|
|
|
|
|
case BLK_ZONE_TYPE_SEQWRITE_PREF:
|
|
|
|
|
if (idx)
|
|
|
|
|
zd->nr_zones[ZONEFS_ZTYPE_SEQ]++;
|
|
|
|
|
break;
|
|
|
|
|
default:
|
|
|
|
|
zonefs_err(zd->sb, "Unsupported zone type 0x%x\n",
|
|
|
|
|
zone->type);
|
|
|
|
|
return -EIO;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
memcpy(&zd->zones[idx], zone, sizeof(struct blk_zone));
|
|
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int zonefs_get_zone_info(struct zonefs_zone_data *zd)
|
|
|
|
|
{
|
|
|
|
|
struct block_device *bdev = zd->sb->s_bdev;
|
|
|
|
|
int ret;
|
|
|
|
|
|
2022-07-06 15:03:45 +08:00
|
|
|
zd->zones = kvcalloc(bdev_nr_zones(bdev), sizeof(struct blk_zone),
|
|
|
|
|
GFP_KERNEL);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
if (!zd->zones)
|
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
|
|
/* Get zones information from the device */
|
|
|
|
|
ret = blkdev_report_zones(bdev, 0, BLK_ALL_ZONES,
|
|
|
|
|
zonefs_get_zone_info_cb, zd);
|
|
|
|
|
if (ret < 0) {
|
|
|
|
|
zonefs_err(zd->sb, "Zone report failed %d\n", ret);
|
|
|
|
|
return ret;
|
|
|
|
|
}
|
|
|
|
|
|
2022-07-06 15:03:45 +08:00
|
|
|
if (ret != bdev_nr_zones(bdev)) {
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
zonefs_err(zd->sb, "Invalid zone report (%d/%u zones)\n",
|
2022-07-06 15:03:45 +08:00
|
|
|
ret, bdev_nr_zones(bdev));
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
return -EIO;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static inline void zonefs_cleanup_zone_info(struct zonefs_zone_data *zd)
|
|
|
|
|
{
|
|
|
|
|
kvfree(zd->zones);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Read super block information from the device.
|
|
|
|
|
*/
|
|
|
|
|
static int zonefs_read_super(struct super_block *sb)
|
|
|
|
|
{
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
|
|
|
|
|
struct zonefs_super *super;
|
|
|
|
|
u32 crc, stored_crc;
|
|
|
|
|
struct page *page;
|
|
|
|
|
struct bio_vec bio_vec;
|
|
|
|
|
struct bio bio;
|
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
|
|
page = alloc_page(GFP_KERNEL);
|
|
|
|
|
if (!page)
|
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
2022-01-24 17:11:06 +08:00
|
|
|
bio_init(&bio, sb->s_bdev, &bio_vec, 1, REQ_OP_READ);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
bio.bi_iter.bi_sector = 0;
|
|
|
|
|
bio_add_page(&bio, page, PAGE_SIZE, 0);
|
|
|
|
|
|
|
|
|
|
ret = submit_bio_wait(&bio);
|
|
|
|
|
if (ret)
|
|
|
|
|
goto free_page;
|
|
|
|
|
|
2022-07-05 22:22:02 +08:00
|
|
|
super = page_address(page);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
ret = -EINVAL;
|
|
|
|
|
if (le32_to_cpu(super->s_magic) != ZONEFS_MAGIC)
|
2022-07-05 22:22:02 +08:00
|
|
|
goto free_page;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
stored_crc = le32_to_cpu(super->s_crc);
|
|
|
|
|
super->s_crc = 0;
|
|
|
|
|
crc = crc32(~0U, (unsigned char *)super, sizeof(struct zonefs_super));
|
|
|
|
|
if (crc != stored_crc) {
|
|
|
|
|
zonefs_err(sb, "Invalid checksum (Expected 0x%08x, got 0x%08x)",
|
|
|
|
|
crc, stored_crc);
|
2022-07-05 22:22:02 +08:00
|
|
|
goto free_page;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
sbi->s_features = le64_to_cpu(super->s_features);
|
|
|
|
|
if (sbi->s_features & ~ZONEFS_F_DEFINED_FEATURES) {
|
|
|
|
|
zonefs_err(sb, "Unknown features set 0x%llx\n",
|
|
|
|
|
sbi->s_features);
|
2022-07-05 22:22:02 +08:00
|
|
|
goto free_page;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (sbi->s_features & ZONEFS_F_UID) {
|
|
|
|
|
sbi->s_uid = make_kuid(current_user_ns(),
|
|
|
|
|
le32_to_cpu(super->s_uid));
|
|
|
|
|
if (!uid_valid(sbi->s_uid)) {
|
|
|
|
|
zonefs_err(sb, "Invalid UID feature\n");
|
2022-07-05 22:22:02 +08:00
|
|
|
goto free_page;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (sbi->s_features & ZONEFS_F_GID) {
|
|
|
|
|
sbi->s_gid = make_kgid(current_user_ns(),
|
|
|
|
|
le32_to_cpu(super->s_gid));
|
|
|
|
|
if (!gid_valid(sbi->s_gid)) {
|
|
|
|
|
zonefs_err(sb, "Invalid GID feature\n");
|
2022-07-05 22:22:02 +08:00
|
|
|
goto free_page;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (sbi->s_features & ZONEFS_F_PERM)
|
|
|
|
|
sbi->s_perm = le32_to_cpu(super->s_perm);
|
|
|
|
|
|
|
|
|
|
if (memchr_inv(super->s_reserved, 0, sizeof(super->s_reserved))) {
|
|
|
|
|
zonefs_err(sb, "Reserved area is being used\n");
|
2022-07-05 22:22:02 +08:00
|
|
|
goto free_page;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
|
2020-04-23 23:32:11 +08:00
|
|
|
import_uuid(&sbi->s_uuid, super->s_uuid);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
ret = 0;
|
|
|
|
|
|
|
|
|
|
free_page:
|
|
|
|
|
__free_page(page);
|
|
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
|
}
|
|
|
|
|
|
2022-11-25 08:39:33 +08:00
|
|
|
static const struct super_operations zonefs_sops = {
|
|
|
|
|
.alloc_inode = zonefs_alloc_inode,
|
|
|
|
|
.free_inode = zonefs_free_inode,
|
|
|
|
|
.statfs = zonefs_statfs,
|
|
|
|
|
.remount_fs = zonefs_remount,
|
|
|
|
|
.show_options = zonefs_show_options,
|
|
|
|
|
};
|
|
|
|
|
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
/*
|
|
|
|
|
* Check that the device is zoned. If it is, get the list of zones and create
|
|
|
|
|
* sub-directories and files according to the device zone configuration and
|
|
|
|
|
* format options.
|
|
|
|
|
*/
|
|
|
|
|
static int zonefs_fill_super(struct super_block *sb, void *data, int silent)
|
|
|
|
|
{
|
|
|
|
|
struct zonefs_zone_data zd;
|
|
|
|
|
struct zonefs_sb_info *sbi;
|
|
|
|
|
struct inode *inode;
|
|
|
|
|
enum zonefs_ztype t;
|
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
|
|
if (!bdev_is_zoned(sb->s_bdev)) {
|
|
|
|
|
zonefs_err(sb, "Not a zoned block device\n");
|
|
|
|
|
return -EINVAL;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Initialize super block information: the maximum file size is updated
|
|
|
|
|
* when the zone files are created so that the format option
|
|
|
|
|
* ZONEFS_F_AGGRCNV which increases the maximum file size of a file
|
|
|
|
|
* beyond the zone size is taken into account.
|
|
|
|
|
*/
|
|
|
|
|
sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
|
|
|
|
|
if (!sbi)
|
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
|
|
spin_lock_init(&sbi->s_lock);
|
|
|
|
|
sb->s_fs_info = sbi;
|
|
|
|
|
sb->s_magic = ZONEFS_MAGIC;
|
|
|
|
|
sb->s_maxbytes = 0;
|
|
|
|
|
sb->s_op = &zonefs_sops;
|
|
|
|
|
sb->s_time_gran = 1;
|
|
|
|
|
|
|
|
|
|
/*
|
2021-01-28 12:47:31 +08:00
|
|
|
* The block size is set to the device zone write granularity to ensure
|
|
|
|
|
* that write operations are always aligned according to the device
|
|
|
|
|
* interface constraints.
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
*/
|
2021-01-28 12:47:31 +08:00
|
|
|
sb_set_blocksize(sb, bdev_zone_write_granularity(sb->s_bdev));
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
sbi->s_zone_sectors_shift = ilog2(bdev_zone_sectors(sb->s_bdev));
|
|
|
|
|
sbi->s_uid = GLOBAL_ROOT_UID;
|
|
|
|
|
sbi->s_gid = GLOBAL_ROOT_GID;
|
|
|
|
|
sbi->s_perm = 0640;
|
|
|
|
|
sbi->s_mount_opts = ZONEFS_MNTOPT_ERRORS_RO;
|
2022-04-12 16:00:13 +08:00
|
|
|
|
|
|
|
|
atomic_set(&sbi->s_wro_seq_files, 0);
|
|
|
|
|
sbi->s_max_wro_seq_files = bdev_max_open_zones(sb->s_bdev);
|
2022-04-12 17:54:39 +08:00
|
|
|
atomic_set(&sbi->s_active_seq_files, 0);
|
|
|
|
|
sbi->s_max_active_seq_files = bdev_max_active_zones(sb->s_bdev);
|
|
|
|
|
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
ret = zonefs_read_super(sb);
|
|
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
|
|
ret = zonefs_parse_options(sb, data);
|
|
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
|
|
memset(&zd, 0, sizeof(struct zonefs_zone_data));
|
|
|
|
|
zd.sb = sb;
|
|
|
|
|
ret = zonefs_get_zone_info(&zd);
|
|
|
|
|
if (ret)
|
|
|
|
|
goto cleanup;
|
|
|
|
|
|
2022-04-12 15:25:34 +08:00
|
|
|
ret = zonefs_sysfs_register(sb);
|
|
|
|
|
if (ret)
|
|
|
|
|
goto cleanup;
|
|
|
|
|
|
2022-07-06 15:03:45 +08:00
|
|
|
zonefs_info(sb, "Mounting %u zones", bdev_nr_zones(sb->s_bdev));
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
2022-06-02 22:16:57 +08:00
|
|
|
if (!sbi->s_max_wro_seq_files &&
|
2022-06-02 20:33:25 +08:00
|
|
|
!sbi->s_max_active_seq_files &&
|
2022-06-02 22:16:57 +08:00
|
|
|
sbi->s_mount_opts & ZONEFS_MNTOPT_EXPLICIT_OPEN) {
|
2022-06-02 20:33:25 +08:00
|
|
|
zonefs_info(sb,
|
|
|
|
|
"No open and active zone limits. Ignoring explicit_open mount option\n");
|
2022-06-02 22:16:57 +08:00
|
|
|
sbi->s_mount_opts &= ~ZONEFS_MNTOPT_EXPLICIT_OPEN;
|
|
|
|
|
}
|
|
|
|
|
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
/* Create root directory inode */
|
|
|
|
|
ret = -ENOMEM;
|
|
|
|
|
inode = new_inode(sb);
|
|
|
|
|
if (!inode)
|
|
|
|
|
goto cleanup;
|
|
|
|
|
|
2022-07-06 15:03:45 +08:00
|
|
|
inode->i_ino = bdev_nr_zones(sb->s_bdev);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
inode->i_mode = S_IFDIR | 0555;
|
|
|
|
|
inode->i_ctime = inode->i_mtime = inode->i_atime = current_time(inode);
|
|
|
|
|
inode->i_op = &zonefs_dir_inode_operations;
|
|
|
|
|
inode->i_fop = &simple_dir_operations;
|
|
|
|
|
set_nlink(inode, 2);
|
|
|
|
|
|
|
|
|
|
sb->s_root = d_make_root(inode);
|
|
|
|
|
if (!sb->s_root)
|
|
|
|
|
goto cleanup;
|
|
|
|
|
|
|
|
|
|
/* Create and populate files in zone groups directories */
|
|
|
|
|
for (t = 0; t < ZONEFS_ZTYPE_MAX; t++) {
|
|
|
|
|
ret = zonefs_create_zgroup(&zd, t);
|
|
|
|
|
if (ret)
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
cleanup:
|
|
|
|
|
zonefs_cleanup_zone_info(&zd);
|
|
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static struct dentry *zonefs_mount(struct file_system_type *fs_type,
|
|
|
|
|
int flags, const char *dev_name, void *data)
|
|
|
|
|
{
|
|
|
|
|
return mount_bdev(fs_type, flags, dev_name, data, zonefs_fill_super);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void zonefs_kill_super(struct super_block *sb)
|
|
|
|
|
{
|
|
|
|
|
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
|
|
|
|
|
|
|
|
|
|
if (sb->s_root)
|
|
|
|
|
d_genocide(sb->s_root);
|
2022-04-12 15:25:34 +08:00
|
|
|
|
|
|
|
|
zonefs_sysfs_unregister(sb);
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
kill_block_super(sb);
|
|
|
|
|
kfree(sbi);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* File system definition and registration.
|
|
|
|
|
*/
|
|
|
|
|
static struct file_system_type zonefs_type = {
|
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
|
.name = "zonefs",
|
|
|
|
|
.mount = zonefs_mount,
|
|
|
|
|
.kill_sb = zonefs_kill_super,
|
|
|
|
|
.fs_flags = FS_REQUIRES_DEV,
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static int __init zonefs_init_inodecache(void)
|
|
|
|
|
{
|
|
|
|
|
zonefs_inode_cachep = kmem_cache_create("zonefs_inode_cache",
|
|
|
|
|
sizeof(struct zonefs_inode_info), 0,
|
|
|
|
|
(SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD | SLAB_ACCOUNT),
|
|
|
|
|
NULL);
|
|
|
|
|
if (zonefs_inode_cachep == NULL)
|
|
|
|
|
return -ENOMEM;
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void zonefs_destroy_inodecache(void)
|
|
|
|
|
{
|
|
|
|
|
/*
|
|
|
|
|
* Make sure all delayed rcu free inodes are flushed before we
|
|
|
|
|
* destroy the inode cache.
|
|
|
|
|
*/
|
|
|
|
|
rcu_barrier();
|
|
|
|
|
kmem_cache_destroy(zonefs_inode_cachep);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int __init zonefs_init(void)
|
|
|
|
|
{
|
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
|
|
BUILD_BUG_ON(sizeof(struct zonefs_super) != ZONEFS_SUPER_SIZE);
|
|
|
|
|
|
|
|
|
|
ret = zonefs_init_inodecache();
|
|
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
|
2022-11-20 18:57:59 +08:00
|
|
|
ret = zonefs_sysfs_init();
|
2022-04-12 15:25:34 +08:00
|
|
|
if (ret)
|
|
|
|
|
goto destroy_inodecache;
|
|
|
|
|
|
2022-11-20 18:57:59 +08:00
|
|
|
ret = register_filesystem(&zonefs_type);
|
2022-04-12 15:25:34 +08:00
|
|
|
if (ret)
|
2022-11-20 18:57:59 +08:00
|
|
|
goto sysfs_exit;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
|
|
|
|
|
return 0;
|
2022-04-12 15:25:34 +08:00
|
|
|
|
2022-11-20 18:57:59 +08:00
|
|
|
sysfs_exit:
|
|
|
|
|
zonefs_sysfs_exit();
|
2022-04-12 15:25:34 +08:00
|
|
|
destroy_inodecache:
|
|
|
|
|
zonefs_destroy_inodecache();
|
|
|
|
|
|
|
|
|
|
return ret;
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void __exit zonefs_exit(void)
|
|
|
|
|
{
|
2022-11-20 18:57:59 +08:00
|
|
|
unregister_filesystem(&zonefs_type);
|
2022-04-12 15:25:34 +08:00
|
|
|
zonefs_sysfs_exit();
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
zonefs_destroy_inodecache();
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
MODULE_AUTHOR("Damien Le Moal");
|
|
|
|
|
MODULE_DESCRIPTION("Zone file system for zoned block devices");
|
|
|
|
|
MODULE_LICENSE("GPL");
|
2021-12-17 14:15:45 +08:00
|
|
|
MODULE_ALIAS_FS("zonefs");
|
fs: New zonefs file system
zonefs is a very simple file system exposing each zone of a zoned block
device as a file. Unlike a regular file system with zoned block device
support (e.g. f2fs), zonefs does not hide the sequential write
constraint of zoned block devices to the user. Files representing
sequential write zones of the device must be written sequentially
starting from the end of the file (append only writes).
As such, zonefs is in essence closer to a raw block device access
interface than to a full featured POSIX file system. The goal of zonefs
is to simplify the implementation of zoned block device support in
applications by replacing raw block device file accesses with a richer
file API, avoiding relying on direct block device file ioctls which may
be more obscure to developers. One example of this approach is the
implementation of LSM (log-structured merge) tree structures (such as
used in RocksDB and LevelDB) on zoned block devices by allowing SSTables
to be stored in a zone file similarly to a regular file system rather
than as a range of sectors of a zoned device. The introduction of the
higher level construct "one file is one zone" can help reducing the
amount of changes needed in the application as well as introducing
support for different application programming languages.
Zonefs on-disk metadata is reduced to an immutable super block to
persistently store a magic number and optional feature flags and
values. On mount, zonefs uses blkdev_report_zones() to obtain the device
zone configuration and populates the mount point with a static file tree
solely based on this information. E.g. file sizes come from the device
zone type and write pointer offset managed by the device itself.
The zone files created on mount have the following characteristics.
1) Files representing zones of the same type are grouped together
under a common sub-directory:
* For conventional zones, the sub-directory "cnv" is used.
* For sequential write zones, the sub-directory "seq" is used.
These two directories are the only directories that exist in zonefs.
Users cannot create other directories and cannot rename nor delete
the "cnv" and "seq" sub-directories.
2) The name of zone files is the number of the file within the zone
type sub-directory, in order of increasing zone start sector.
3) The size of conventional zone files is fixed to the device zone size.
Conventional zone files cannot be truncated.
4) The size of sequential zone files represent the file's zone write
pointer position relative to the zone start sector. Truncating these
files is allowed only down to 0, in which case, the zone is reset to
rewind the zone write pointer position to the start of the zone, or
up to the zone size, in which case the file's zone is transitioned
to the FULL state (finish zone operation).
5) All read and write operations to files are not allowed beyond the
file zone size. Any access exceeding the zone size is failed with
the -EFBIG error.
6) Creating, deleting, renaming or modifying any attribute of files and
sub-directories is not allowed.
7) There are no restrictions on the type of read and write operations
that can be issued to conventional zone files. Buffered, direct and
mmap read & write operations are accepted. For sequential zone files,
there are no restrictions on read operations, but all write
operations must be direct IO append writes. mmap write of sequential
files is not allowed.
Several optional features of zonefs can be enabled at format time.
* Conventional zone aggregation: ranges of contiguous conventional
zones can be aggregated into a single larger file instead of the
default one file per zone.
* File ownership: The owner UID and GID of zone files is by default 0
(root) but can be changed to any valid UID/GID.
* File access permissions: the default 640 access permissions can be
changed.
The mkzonefs tool is used to format zoned block devices for use with
zonefs. This tool is available on Github at:
git@github.com:damien-lemoal/zonefs-tools.git.
zonefs-tools also includes a test suite which can be run against any
zoned block device, including null_blk block device created with zoned
mode.
Example: the following formats a 15TB host-managed SMR HDD with 256 MB
zones with the conventional zones aggregation feature enabled.
$ sudo mkzonefs -o aggr_cnv /dev/sdX
$ sudo mount -t zonefs /dev/sdX /mnt
$ ls -l /mnt/
total 0
dr-xr-xr-x 2 root root 1 Nov 25 13:23 cnv
dr-xr-xr-x 2 root root 55356 Nov 25 13:23 seq
The size of the zone files sub-directories indicate the number of files
existing for each type of zones. In this example, there is only one
conventional zone file (all conventional zones are aggregated under a
single file).
$ ls -l /mnt/cnv
total 137101312
-rw-r----- 1 root root 140391743488 Nov 25 13:23 0
This aggregated conventional zone file can be used as a regular file.
$ sudo mkfs.ext4 /mnt/cnv/0
$ sudo mount -o loop /mnt/cnv/0 /data
The "seq" sub-directory grouping files for sequential write zones has
in this example 55356 zones.
$ ls -lv /mnt/seq
total 14511243264
-rw-r----- 1 root root 0 Nov 25 13:23 0
-rw-r----- 1 root root 0 Nov 25 13:23 1
-rw-r----- 1 root root 0 Nov 25 13:23 2
...
-rw-r----- 1 root root 0 Nov 25 13:23 55354
-rw-r----- 1 root root 0 Nov 25 13:23 55355
For sequential write zone files, the file size changes as data is
appended at the end of the file, similarly to any regular file system.
$ dd if=/dev/zero of=/mnt/seq/0 bs=4K count=1 conv=notrunc oflag=direct
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000452219 s, 9.1 MB/s
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/seq/0
The written file can be truncated to the zone size, preventing any
further write operation.
$ truncate -s 268435456 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 268435456 Nov 25 13:49 /mnt/seq/0
Truncation to 0 size allows freeing the file zone storage space and
restart append-writes to the file.
$ truncate -s 0 /mnt/seq/0
$ ls -l /mnt/seq/0
-rw-r----- 1 root root 0 Nov 25 13:49 /mnt/seq/0
Since files are statically mapped to zones on the disk, the number of
blocks of a file as reported by stat() and fstat() indicates the size
of the file zone.
$ stat /mnt/seq/0
File: /mnt/seq/0
Size: 0 Blocks: 524288 IO Block: 4096 regular empty file
Device: 870h/2160d Inode: 50431 Links: 1
Access: (0640/-rw-r-----) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2019-11-25 13:23:57.048971997 +0900
Modify: 2019-11-25 13:52:25.553805765 +0900
Change: 2019-11-25 13:52:25.553805765 +0900
Birth: -
The number of blocks of the file ("Blocks") in units of 512B blocks
gives the maximum file size of 524288 * 512 B = 256 MB, corresponding
to the device zone size in this example. Of note is that the "IO block"
field always indicates the minimum IO size for writes and corresponds
to the device physical sector size.
This code contains contributions from:
* Johannes Thumshirn <jthumshirn@suse.de>,
* Darrick J. Wong <darrick.wong@oracle.com>,
* Christoph Hellwig <hch@lst.de>,
* Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> and
* Ting Yao <tingyao@hust.edu.cn>.
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
2019-12-25 15:07:44 +08:00
|
|
|
module_init(zonefs_init);
|
|
|
|
|
module_exit(zonefs_exit);
|
