2007-06-12 09:07:21 -04:00
|
|
|
/*
|
|
|
|
* Copyright (C) 2007 Oracle. All rights reserved.
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU General Public
|
|
|
|
* License v2 as published by the Free Software Foundation.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
|
|
* General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public
|
|
|
|
* License along with this program; if not, write to the
|
|
|
|
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
|
|
|
|
* Boston, MA 021110-1307, USA.
|
|
|
|
*/
|
|
|
|
|
2007-08-29 09:11:44 -04:00
|
|
|
#include <linux/blkdev.h>
|
2007-03-21 11:12:56 -04:00
|
|
|
#include <linux/module.h>
|
2007-03-22 12:13:20 -04:00
|
|
|
#include <linux/buffer_head.h>
|
2007-03-21 11:12:56 -04:00
|
|
|
#include <linux/fs.h>
|
|
|
|
#include <linux/pagemap.h>
|
|
|
|
#include <linux/highmem.h>
|
|
|
|
#include <linux/time.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/smp_lock.h>
|
|
|
|
#include <linux/backing-dev.h>
|
2007-08-29 09:11:44 -04:00
|
|
|
#include <linux/mount.h>
|
2007-03-26 16:00:06 -04:00
|
|
|
#include <linux/mpage.h>
|
2007-03-29 11:56:46 -04:00
|
|
|
#include <linux/swap.h>
|
|
|
|
#include <linux/writeback.h>
|
2007-04-19 21:01:03 -04:00
|
|
|
#include <linux/statfs.h>
|
2007-06-08 15:33:54 -04:00
|
|
|
#include <linux/compat.h>
|
2007-08-29 09:11:44 -04:00
|
|
|
#include <linux/parser.h>
|
2007-12-17 20:14:04 -05:00
|
|
|
#include <linux/ctype.h>
|
2007-12-18 16:15:09 -05:00
|
|
|
#include <linux/namei.h>
|
2008-03-24 15:02:04 -04:00
|
|
|
#include <linux/miscdevice.h>
|
2008-11-20 10:22:27 -05:00
|
|
|
#include <linux/version.h>
|
|
|
|
#include "compat.h"
|
2007-03-21 11:12:56 -04:00
|
|
|
#include "ctree.h"
|
2007-03-22 12:13:20 -04:00
|
|
|
#include "disk-io.h"
|
2007-03-23 10:01:08 -04:00
|
|
|
#include "transaction.h"
|
2007-04-02 10:50:19 -04:00
|
|
|
#include "btrfs_inode.h"
|
2007-04-10 09:27:04 -04:00
|
|
|
#include "ioctl.h"
|
2007-05-24 13:35:57 -04:00
|
|
|
#include "print-tree.h"
|
2007-11-16 11:45:54 -05:00
|
|
|
#include "xattr.h"
|
2008-03-24 15:02:07 -04:00
|
|
|
#include "volumes.h"
|
2008-07-23 12:12:13 -04:00
|
|
|
#include "version.h"
|
2008-07-20 16:31:56 -04:00
|
|
|
#include "export.h"
|
Btrfs: Add zlib compression support
This is a large change for adding compression on reading and writing,
both for inline and regular extents. It does some fairly large
surgery to the writeback paths.
Compression is off by default and enabled by mount -o compress. Even
when the -o compress mount option is not used, it is possible to read
compressed extents off the disk.
If compression for a given set of pages fails to make them smaller, the
file is flagged to avoid future compression attempts later.
* While finding delalloc extents, the pages are locked before being sent down
to the delalloc handler. This allows the delalloc handler to do complex things
such as cleaning the pages, marking them writeback and starting IO on their
behalf.
* Inline extents are inserted at delalloc time now. This allows us to compress
the data before inserting the inline extent, and it allows us to insert
an inline extent that spans multiple pages.
* All of the in-memory extent representations (extent_map.c, ordered-data.c etc)
are changed to record both an in-memory size and an on disk size, as well
as a flag for compression.
From a disk format point of view, the extent pointers in the file are changed
to record the on disk size of a given extent and some encoding flags.
Space in the disk format is allocated for compression encoding, as well
as encryption and a generic 'other' field. Neither the encryption or the
'other' field are currently used.
In order to limit the amount of data read for a single random read in the
file, the size of a compressed extent is limited to 128k. This is a
software only limit, the disk format supports u64 sized compressed extents.
In order to limit the ram consumed while processing extents, the uncompressed
size of a compressed extent is limited to 256k. This is a software only limit
and will be subject to tuning later.
Checksumming is still done on compressed extents, and it is done on the
uncompressed version of the data. This way additional encodings can be
layered on without having to figure out which encoding to checksum.
Compression happens at delalloc time, which is basically singled threaded because
it is usually done by a single pdflush thread. This makes it tricky to
spread the compression load across all the cpus on the box. We'll have to
look at parallel pdflush walks of dirty inodes at a later time.
Decompression is hooked into readpages and it does spread across CPUs nicely.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-10-29 14:49:59 -04:00
|
|
|
#include "compression.h"
|
2007-03-21 11:12:56 -04:00
|
|
|
|
2007-10-15 16:14:19 -04:00
|
|
|
#define BTRFS_SUPER_MAGIC 0x9123683E
|
2007-03-29 15:15:27 -04:00
|
|
|
|
2007-06-12 06:35:45 -04:00
|
|
|
static struct super_operations btrfs_super_ops;
|
2007-03-29 11:56:46 -04:00
|
|
|
|
2009-01-05 21:25:51 -05:00
|
|
|
static void btrfs_put_super(struct super_block *sb)
|
2007-04-17 13:26:50 -04:00
|
|
|
{
|
2007-06-12 06:35:45 -04:00
|
|
|
struct btrfs_root *root = btrfs_sb(sb);
|
2007-04-17 13:26:50 -04:00
|
|
|
int ret;
|
|
|
|
|
2007-06-12 06:35:45 -04:00
|
|
|
ret = close_ctree(root);
|
|
|
|
sb->s_fs_info = NULL;
|
2007-03-29 11:56:46 -04:00
|
|
|
}
|
|
|
|
|
2007-08-29 09:11:44 -04:00
|
|
|
enum {
|
2008-06-10 10:40:46 -04:00
|
|
|
Opt_degraded, Opt_subvol, Opt_device, Opt_nodatasum, Opt_nodatacow,
|
2008-05-13 13:46:40 -04:00
|
|
|
Opt_max_extent, Opt_max_inline, Opt_alloc_start, Opt_nobarrier,
|
Btrfs: Add zlib compression support
This is a large change for adding compression on reading and writing,
both for inline and regular extents. It does some fairly large
surgery to the writeback paths.
Compression is off by default and enabled by mount -o compress. Even
when the -o compress mount option is not used, it is possible to read
compressed extents off the disk.
If compression for a given set of pages fails to make them smaller, the
file is flagged to avoid future compression attempts later.
* While finding delalloc extents, the pages are locked before being sent down
to the delalloc handler. This allows the delalloc handler to do complex things
such as cleaning the pages, marking them writeback and starting IO on their
behalf.
* Inline extents are inserted at delalloc time now. This allows us to compress
the data before inserting the inline extent, and it allows us to insert
an inline extent that spans multiple pages.
* All of the in-memory extent representations (extent_map.c, ordered-data.c etc)
are changed to record both an in-memory size and an on disk size, as well
as a flag for compression.
From a disk format point of view, the extent pointers in the file are changed
to record the on disk size of a given extent and some encoding flags.
Space in the disk format is allocated for compression encoding, as well
as encryption and a generic 'other' field. Neither the encryption or the
'other' field are currently used.
In order to limit the amount of data read for a single random read in the
file, the size of a compressed extent is limited to 128k. This is a
software only limit, the disk format supports u64 sized compressed extents.
In order to limit the ram consumed while processing extents, the uncompressed
size of a compressed extent is limited to 256k. This is a software only limit
and will be subject to tuning later.
Checksumming is still done on compressed extents, and it is done on the
uncompressed version of the data. This way additional encodings can be
layered on without having to figure out which encoding to checksum.
Compression happens at delalloc time, which is basically singled threaded because
it is usually done by a single pdflush thread. This makes it tricky to
spread the compression load across all the cpus on the box. We'll have to
look at parallel pdflush walks of dirty inodes at a later time.
Decompression is hooked into readpages and it does spread across CPUs nicely.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-10-29 14:49:59 -04:00
|
|
|
Opt_ssd, Opt_thread_pool, Opt_noacl, Opt_compress, Opt_err,
|
2007-08-29 09:11:44 -04:00
|
|
|
};
|
|
|
|
|
|
|
|
static match_table_t tokens = {
|
2008-05-13 13:46:40 -04:00
|
|
|
{Opt_degraded, "degraded"},
|
2007-08-29 09:11:44 -04:00
|
|
|
{Opt_subvol, "subvol=%s"},
|
2008-06-10 10:40:46 -04:00
|
|
|
{Opt_device, "device=%s"},
|
2007-12-14 15:30:32 -05:00
|
|
|
{Opt_nodatasum, "nodatasum"},
|
2007-12-17 20:14:01 -05:00
|
|
|
{Opt_nodatacow, "nodatacow"},
|
2008-01-09 09:23:21 -05:00
|
|
|
{Opt_nobarrier, "nobarrier"},
|
2007-12-17 20:14:04 -05:00
|
|
|
{Opt_max_extent, "max_extent=%s"},
|
2008-01-29 16:03:38 -05:00
|
|
|
{Opt_max_inline, "max_inline=%s"},
|
2008-01-02 10:01:11 -05:00
|
|
|
{Opt_alloc_start, "alloc_start=%s"},
|
2008-06-11 21:47:56 -04:00
|
|
|
{Opt_thread_pool, "thread_pool=%d"},
|
Btrfs: Add zlib compression support
This is a large change for adding compression on reading and writing,
both for inline and regular extents. It does some fairly large
surgery to the writeback paths.
Compression is off by default and enabled by mount -o compress. Even
when the -o compress mount option is not used, it is possible to read
compressed extents off the disk.
If compression for a given set of pages fails to make them smaller, the
file is flagged to avoid future compression attempts later.
* While finding delalloc extents, the pages are locked before being sent down
to the delalloc handler. This allows the delalloc handler to do complex things
such as cleaning the pages, marking them writeback and starting IO on their
behalf.
* Inline extents are inserted at delalloc time now. This allows us to compress
the data before inserting the inline extent, and it allows us to insert
an inline extent that spans multiple pages.
* All of the in-memory extent representations (extent_map.c, ordered-data.c etc)
are changed to record both an in-memory size and an on disk size, as well
as a flag for compression.
From a disk format point of view, the extent pointers in the file are changed
to record the on disk size of a given extent and some encoding flags.
Space in the disk format is allocated for compression encoding, as well
as encryption and a generic 'other' field. Neither the encryption or the
'other' field are currently used.
In order to limit the amount of data read for a single random read in the
file, the size of a compressed extent is limited to 128k. This is a
software only limit, the disk format supports u64 sized compressed extents.
In order to limit the ram consumed while processing extents, the uncompressed
size of a compressed extent is limited to 256k. This is a software only limit
and will be subject to tuning later.
Checksumming is still done on compressed extents, and it is done on the
uncompressed version of the data. This way additional encodings can be
layered on without having to figure out which encoding to checksum.
Compression happens at delalloc time, which is basically singled threaded because
it is usually done by a single pdflush thread. This makes it tricky to
spread the compression load across all the cpus on the box. We'll have to
look at parallel pdflush walks of dirty inodes at a later time.
Decompression is hooked into readpages and it does spread across CPUs nicely.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-10-29 14:49:59 -04:00
|
|
|
{Opt_compress, "compress"},
|
2008-01-18 10:54:22 -05:00
|
|
|
{Opt_ssd, "ssd"},
|
2008-07-24 12:16:36 -04:00
|
|
|
{Opt_noacl, "noacl"},
|
|
|
|
{Opt_err, NULL},
|
2007-08-29 09:11:44 -04:00
|
|
|
};
|
|
|
|
|
2007-12-21 16:27:24 -05:00
|
|
|
u64 btrfs_parse_size(char *str)
|
2007-12-17 20:14:04 -05:00
|
|
|
{
|
2007-12-21 16:27:24 -05:00
|
|
|
u64 res;
|
2007-12-17 20:14:04 -05:00
|
|
|
int mult = 1;
|
|
|
|
char *end;
|
|
|
|
char last;
|
|
|
|
|
|
|
|
res = simple_strtoul(str, &end, 10);
|
|
|
|
|
|
|
|
last = end[0];
|
|
|
|
if (isalpha(last)) {
|
|
|
|
last = tolower(last);
|
|
|
|
switch (last) {
|
|
|
|
case 'g':
|
|
|
|
mult *= 1024;
|
|
|
|
case 'm':
|
|
|
|
mult *= 1024;
|
|
|
|
case 'k':
|
|
|
|
mult *= 1024;
|
|
|
|
}
|
|
|
|
res = res * mult;
|
|
|
|
}
|
|
|
|
return res;
|
|
|
|
}
|
|
|
|
|
2008-06-10 10:40:29 -04:00
|
|
|
/*
|
|
|
|
* Regular mount options parser. Everything that is needed only when
|
|
|
|
* reading in a new superblock is parsed here.
|
|
|
|
*/
|
|
|
|
int btrfs_parse_options(struct btrfs_root *root, char *options)
|
2007-08-29 09:11:44 -04:00
|
|
|
{
|
2008-06-10 10:40:29 -04:00
|
|
|
struct btrfs_fs_info *info = root->fs_info;
|
2007-08-29 09:11:44 -04:00
|
|
|
substring_t args[MAX_OPT_ARGS];
|
2008-06-10 10:40:29 -04:00
|
|
|
char *p, *num;
|
2008-06-11 21:47:56 -04:00
|
|
|
int intarg;
|
2007-12-14 15:30:32 -05:00
|
|
|
|
2007-08-29 09:11:44 -04:00
|
|
|
if (!options)
|
2008-06-10 10:40:29 -04:00
|
|
|
return 0;
|
2007-08-29 09:11:44 -04:00
|
|
|
|
2007-12-17 20:14:01 -05:00
|
|
|
/*
|
|
|
|
* strsep changes the string, duplicate it because parse_options
|
|
|
|
* gets called twice
|
|
|
|
*/
|
|
|
|
options = kstrdup(options, GFP_NOFS);
|
|
|
|
if (!options)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
|
2008-06-10 10:40:29 -04:00
|
|
|
while ((p = strsep(&options, ",")) != NULL) {
|
2007-08-29 09:11:44 -04:00
|
|
|
int token;
|
|
|
|
if (!*p)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
token = match_token(p, tokens, args);
|
|
|
|
switch (token) {
|
2008-05-13 13:46:40 -04:00
|
|
|
case Opt_degraded:
|
2008-06-10 10:40:29 -04:00
|
|
|
printk(KERN_INFO "btrfs: allowing degraded mounts\n");
|
|
|
|
btrfs_set_opt(info->mount_opt, DEGRADED);
|
2008-05-13 13:46:40 -04:00
|
|
|
break;
|
2007-08-29 09:11:44 -04:00
|
|
|
case Opt_subvol:
|
2008-06-10 10:40:46 -04:00
|
|
|
case Opt_device:
|
2008-06-10 10:40:29 -04:00
|
|
|
/*
|
2008-06-10 10:40:46 -04:00
|
|
|
* These are parsed by btrfs_parse_early_options
|
2008-06-10 10:40:29 -04:00
|
|
|
* and can be happily ignored here.
|
|
|
|
*/
|
2007-12-14 15:30:32 -05:00
|
|
|
break;
|
|
|
|
case Opt_nodatasum:
|
2008-06-10 10:40:29 -04:00
|
|
|
printk(KERN_INFO "btrfs: setting nodatacsum\n");
|
|
|
|
btrfs_set_opt(info->mount_opt, NODATASUM);
|
2007-12-17 20:14:01 -05:00
|
|
|
break;
|
|
|
|
case Opt_nodatacow:
|
2008-06-10 10:40:29 -04:00
|
|
|
printk(KERN_INFO "btrfs: setting nodatacow\n");
|
|
|
|
btrfs_set_opt(info->mount_opt, NODATACOW);
|
|
|
|
btrfs_set_opt(info->mount_opt, NODATASUM);
|
2007-08-29 09:11:44 -04:00
|
|
|
break;
|
Btrfs: Add zlib compression support
This is a large change for adding compression on reading and writing,
both for inline and regular extents. It does some fairly large
surgery to the writeback paths.
Compression is off by default and enabled by mount -o compress. Even
when the -o compress mount option is not used, it is possible to read
compressed extents off the disk.
If compression for a given set of pages fails to make them smaller, the
file is flagged to avoid future compression attempts later.
* While finding delalloc extents, the pages are locked before being sent down
to the delalloc handler. This allows the delalloc handler to do complex things
such as cleaning the pages, marking them writeback and starting IO on their
behalf.
* Inline extents are inserted at delalloc time now. This allows us to compress
the data before inserting the inline extent, and it allows us to insert
an inline extent that spans multiple pages.
* All of the in-memory extent representations (extent_map.c, ordered-data.c etc)
are changed to record both an in-memory size and an on disk size, as well
as a flag for compression.
From a disk format point of view, the extent pointers in the file are changed
to record the on disk size of a given extent and some encoding flags.
Space in the disk format is allocated for compression encoding, as well
as encryption and a generic 'other' field. Neither the encryption or the
'other' field are currently used.
In order to limit the amount of data read for a single random read in the
file, the size of a compressed extent is limited to 128k. This is a
software only limit, the disk format supports u64 sized compressed extents.
In order to limit the ram consumed while processing extents, the uncompressed
size of a compressed extent is limited to 256k. This is a software only limit
and will be subject to tuning later.
Checksumming is still done on compressed extents, and it is done on the
uncompressed version of the data. This way additional encodings can be
layered on without having to figure out which encoding to checksum.
Compression happens at delalloc time, which is basically singled threaded because
it is usually done by a single pdflush thread. This makes it tricky to
spread the compression load across all the cpus on the box. We'll have to
look at parallel pdflush walks of dirty inodes at a later time.
Decompression is hooked into readpages and it does spread across CPUs nicely.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-10-29 14:49:59 -04:00
|
|
|
case Opt_compress:
|
|
|
|
printk(KERN_INFO "btrfs: use compression\n");
|
|
|
|
btrfs_set_opt(info->mount_opt, COMPRESS);
|
|
|
|
break;
|
2008-01-18 10:54:22 -05:00
|
|
|
case Opt_ssd:
|
2008-06-10 10:40:29 -04:00
|
|
|
printk(KERN_INFO "btrfs: use ssd allocation scheme\n");
|
|
|
|
btrfs_set_opt(info->mount_opt, SSD);
|
2008-01-18 10:54:22 -05:00
|
|
|
break;
|
2008-01-09 09:23:21 -05:00
|
|
|
case Opt_nobarrier:
|
2008-06-10 10:40:29 -04:00
|
|
|
printk(KERN_INFO "btrfs: turning off barriers\n");
|
|
|
|
btrfs_set_opt(info->mount_opt, NOBARRIER);
|
2008-01-09 09:23:21 -05:00
|
|
|
break;
|
2008-06-11 21:47:56 -04:00
|
|
|
case Opt_thread_pool:
|
|
|
|
intarg = 0;
|
|
|
|
match_int(&args[0], &intarg);
|
|
|
|
if (intarg) {
|
|
|
|
info->thread_pool_size = intarg;
|
|
|
|
printk(KERN_INFO "btrfs: thread pool %d\n",
|
|
|
|
info->thread_pool_size);
|
|
|
|
}
|
|
|
|
break;
|
2007-12-17 20:14:04 -05:00
|
|
|
case Opt_max_extent:
|
2008-06-10 10:40:29 -04:00
|
|
|
num = match_strdup(&args[0]);
|
|
|
|
if (num) {
|
|
|
|
info->max_extent = btrfs_parse_size(num);
|
|
|
|
kfree(num);
|
|
|
|
|
|
|
|
info->max_extent = max_t(u64,
|
|
|
|
info->max_extent, root->sectorsize);
|
|
|
|
printk(KERN_INFO "btrfs: max_extent at %llu\n",
|
|
|
|
info->max_extent);
|
2007-12-17 20:14:04 -05:00
|
|
|
}
|
|
|
|
break;
|
2008-01-29 16:03:38 -05:00
|
|
|
case Opt_max_inline:
|
2008-06-10 10:40:29 -04:00
|
|
|
num = match_strdup(&args[0]);
|
|
|
|
if (num) {
|
|
|
|
info->max_inline = btrfs_parse_size(num);
|
|
|
|
kfree(num);
|
|
|
|
|
2008-06-11 16:51:38 -04:00
|
|
|
if (info->max_inline) {
|
|
|
|
info->max_inline = max_t(u64,
|
|
|
|
info->max_inline,
|
|
|
|
root->sectorsize);
|
|
|
|
}
|
2008-06-10 10:40:29 -04:00
|
|
|
printk(KERN_INFO "btrfs: max_inline at %llu\n",
|
|
|
|
info->max_inline);
|
2008-01-29 16:03:38 -05:00
|
|
|
}
|
|
|
|
break;
|
2008-01-02 10:01:11 -05:00
|
|
|
case Opt_alloc_start:
|
2008-06-10 10:40:29 -04:00
|
|
|
num = match_strdup(&args[0]);
|
|
|
|
if (num) {
|
|
|
|
info->alloc_start = btrfs_parse_size(num);
|
|
|
|
kfree(num);
|
|
|
|
printk(KERN_INFO
|
|
|
|
"btrfs: allocations start at %llu\n",
|
|
|
|
info->alloc_start);
|
2008-01-02 10:01:11 -05:00
|
|
|
}
|
|
|
|
break;
|
2008-07-24 12:16:36 -04:00
|
|
|
case Opt_noacl:
|
|
|
|
root->fs_info->sb->s_flags &= ~MS_POSIXACL;
|
|
|
|
break;
|
2007-08-29 09:11:44 -04:00
|
|
|
default:
|
2007-12-17 20:14:01 -05:00
|
|
|
break;
|
2007-08-29 09:11:44 -04:00
|
|
|
}
|
|
|
|
}
|
2007-12-17 20:14:01 -05:00
|
|
|
kfree(options);
|
2008-06-10 10:40:29 -04:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Parse mount options that are required early in the mount process.
|
|
|
|
*
|
|
|
|
* All other options will be parsed on much later in the mount process and
|
|
|
|
* only when we need to allocate a new super block.
|
|
|
|
*/
|
2008-12-02 06:36:09 -05:00
|
|
|
static int btrfs_parse_early_options(const char *options, fmode_t flags,
|
2008-06-10 10:40:46 -04:00
|
|
|
void *holder, char **subvol_name,
|
|
|
|
struct btrfs_fs_devices **fs_devices)
|
2008-06-10 10:40:29 -04:00
|
|
|
{
|
|
|
|
substring_t args[MAX_OPT_ARGS];
|
|
|
|
char *opts, *p;
|
|
|
|
int error = 0;
|
|
|
|
|
|
|
|
if (!options)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* strsep changes the string, duplicate it because parse_options
|
|
|
|
* gets called twice
|
|
|
|
*/
|
|
|
|
opts = kstrdup(options, GFP_KERNEL);
|
|
|
|
if (!opts)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
while ((p = strsep(&opts, ",")) != NULL) {
|
|
|
|
int token;
|
|
|
|
if (!*p)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
token = match_token(p, tokens, args);
|
|
|
|
switch (token) {
|
|
|
|
case Opt_subvol:
|
|
|
|
*subvol_name = match_strdup(&args[0]);
|
|
|
|
break;
|
2008-06-10 10:40:46 -04:00
|
|
|
case Opt_device:
|
|
|
|
error = btrfs_scan_one_device(match_strdup(&args[0]),
|
|
|
|
flags, holder, fs_devices);
|
|
|
|
if (error)
|
|
|
|
goto out_free_opts;
|
|
|
|
break;
|
2008-06-10 10:40:29 -04:00
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2008-06-10 10:40:46 -04:00
|
|
|
out_free_opts:
|
2008-06-10 10:40:29 -04:00
|
|
|
kfree(opts);
|
|
|
|
out:
|
|
|
|
/*
|
|
|
|
* If no subvolume name is specified we use the default one. Allocate
|
2008-11-17 21:02:50 -05:00
|
|
|
* a copy of the string "." here so that code later in the
|
2008-06-10 10:40:29 -04:00
|
|
|
* mount path doesn't care if it's the default volume or another one.
|
|
|
|
*/
|
|
|
|
if (!*subvol_name) {
|
2008-11-17 21:02:50 -05:00
|
|
|
*subvol_name = kstrdup(".", GFP_KERNEL);
|
2008-06-10 10:40:29 -04:00
|
|
|
if (!*subvol_name)
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
|
|
|
return error;
|
2007-08-29 09:11:44 -04:00
|
|
|
}
|
|
|
|
|
2009-01-05 21:25:51 -05:00
|
|
|
static int btrfs_fill_super(struct super_block *sb,
|
2008-03-24 15:02:07 -04:00
|
|
|
struct btrfs_fs_devices *fs_devices,
|
2009-01-05 21:25:51 -05:00
|
|
|
void *data, int silent)
|
2007-03-29 11:56:46 -04:00
|
|
|
{
|
2009-01-05 21:25:51 -05:00
|
|
|
struct inode *inode;
|
|
|
|
struct dentry *root_dentry;
|
2007-06-12 06:35:45 -04:00
|
|
|
struct btrfs_super_block *disk_super;
|
|
|
|
struct btrfs_root *tree_root;
|
|
|
|
struct btrfs_inode *bi;
|
|
|
|
int err;
|
2007-04-18 16:15:28 -04:00
|
|
|
|
2007-06-12 06:35:45 -04:00
|
|
|
sb->s_maxbytes = MAX_LFS_FILESIZE;
|
|
|
|
sb->s_magic = BTRFS_SUPER_MAGIC;
|
|
|
|
sb->s_op = &btrfs_super_ops;
|
2008-07-20 16:31:56 -04:00
|
|
|
sb->s_export_op = &btrfs_export_ops;
|
2007-11-16 11:45:54 -05:00
|
|
|
sb->s_xattr = btrfs_xattr_handlers;
|
2007-06-12 06:35:45 -04:00
|
|
|
sb->s_time_gran = 1;
|
2008-07-24 12:16:36 -04:00
|
|
|
sb->s_flags |= MS_POSIXACL;
|
2007-04-18 16:15:28 -04:00
|
|
|
|
2008-05-13 13:46:40 -04:00
|
|
|
tree_root = open_ctree(sb, fs_devices, (char *)data);
|
2007-04-16 09:22:45 -04:00
|
|
|
|
2008-04-01 11:21:34 -04:00
|
|
|
if (IS_ERR(tree_root)) {
|
2007-06-12 06:35:45 -04:00
|
|
|
printk("btrfs: open_ctree failed\n");
|
2008-04-01 11:21:34 -04:00
|
|
|
return PTR_ERR(tree_root);
|
2007-04-18 16:15:28 -04:00
|
|
|
}
|
2007-06-12 06:35:45 -04:00
|
|
|
sb->s_fs_info = tree_root;
|
2007-10-15 16:14:19 -04:00
|
|
|
disk_super = &tree_root->fs_info->super_copy;
|
2008-11-17 21:02:50 -05:00
|
|
|
inode = btrfs_iget_locked(sb, BTRFS_FIRST_FREE_OBJECTID,
|
|
|
|
tree_root->fs_info->fs_root);
|
2007-06-12 06:35:45 -04:00
|
|
|
bi = BTRFS_I(inode);
|
|
|
|
bi->location.objectid = inode->i_ino;
|
|
|
|
bi->location.offset = 0;
|
2008-11-17 21:02:50 -05:00
|
|
|
bi->root = tree_root->fs_info->fs_root;
|
2007-08-27 16:49:44 -04:00
|
|
|
|
2007-06-12 06:35:45 -04:00
|
|
|
btrfs_set_key_type(&bi->location, BTRFS_INODE_ITEM_KEY);
|
2007-04-18 16:15:28 -04:00
|
|
|
|
2007-06-12 06:35:45 -04:00
|
|
|
if (!inode) {
|
2007-04-16 09:22:45 -04:00
|
|
|
err = -ENOMEM;
|
2007-06-12 06:35:45 -04:00
|
|
|
goto fail_close;
|
2007-03-29 15:15:27 -04:00
|
|
|
}
|
2007-06-12 06:35:45 -04:00
|
|
|
if (inode->i_state & I_NEW) {
|
|
|
|
btrfs_read_locked_inode(inode);
|
|
|
|
unlock_new_inode(inode);
|
2007-03-29 15:15:27 -04:00
|
|
|
}
|
|
|
|
|
2007-06-12 06:35:45 -04:00
|
|
|
root_dentry = d_alloc_root(inode);
|
|
|
|
if (!root_dentry) {
|
|
|
|
iput(inode);
|
|
|
|
err = -ENOMEM;
|
|
|
|
goto fail_close;
|
2007-03-29 15:15:27 -04:00
|
|
|
}
|
2008-12-12 10:03:26 -05:00
|
|
|
#if 0
|
2007-08-29 15:47:34 -04:00
|
|
|
/* this does the super kobj at the same time */
|
|
|
|
err = btrfs_sysfs_add_super(tree_root->fs_info);
|
|
|
|
if (err)
|
|
|
|
goto fail_close;
|
2008-12-12 10:03:26 -05:00
|
|
|
#endif
|
2007-08-29 15:47:34 -04:00
|
|
|
|
2007-06-12 06:35:45 -04:00
|
|
|
sb->s_root = root_dentry;
|
2008-02-20 16:11:05 -05:00
|
|
|
|
|
|
|
save_mount_options(sb, data);
|
2007-04-10 16:58:11 -04:00
|
|
|
return 0;
|
2007-06-12 06:35:45 -04:00
|
|
|
|
|
|
|
fail_close:
|
|
|
|
close_ctree(tree_root);
|
|
|
|
return err;
|
2007-04-10 16:58:11 -04:00
|
|
|
}
|
|
|
|
|
2008-06-10 10:07:39 -04:00
|
|
|
int btrfs_sync_fs(struct super_block *sb, int wait)
|
2007-04-10 09:27:04 -04:00
|
|
|
{
|
|
|
|
struct btrfs_trans_handle *trans;
|
2007-06-12 06:35:45 -04:00
|
|
|
struct btrfs_root *root;
|
2007-04-10 09:27:04 -04:00
|
|
|
int ret;
|
2007-06-12 06:35:45 -04:00
|
|
|
root = btrfs_sb(sb);
|
2007-04-10 16:58:11 -04:00
|
|
|
|
2008-11-12 14:34:12 -05:00
|
|
|
if (sb->s_flags & MS_RDONLY)
|
|
|
|
return 0;
|
|
|
|
|
2007-06-12 06:35:45 -04:00
|
|
|
sb->s_dirt = 0;
|
|
|
|
if (!wait) {
|
|
|
|
filemap_flush(root->fs_info->btree_inode->i_mapping);
|
|
|
|
return 0;
|
|
|
|
}
|
2008-11-06 22:02:51 -05:00
|
|
|
|
|
|
|
btrfs_start_delalloc_inodes(root);
|
|
|
|
btrfs_wait_ordered_extents(root, 0);
|
|
|
|
|
2007-08-10 14:06:19 -04:00
|
|
|
btrfs_clean_old_snapshots(root);
|
2007-04-10 09:27:04 -04:00
|
|
|
trans = btrfs_start_transaction(root, 1);
|
|
|
|
ret = btrfs_commit_transaction(trans, root);
|
2007-06-12 06:35:45 -04:00
|
|
|
sb->s_dirt = 0;
|
2007-06-22 14:16:25 -04:00
|
|
|
return ret;
|
2007-04-02 10:50:19 -04:00
|
|
|
}
|
|
|
|
|
2007-06-12 06:35:45 -04:00
|
|
|
static void btrfs_write_super(struct super_block *sb)
|
2007-04-02 10:50:19 -04:00
|
|
|
{
|
2007-06-12 06:35:45 -04:00
|
|
|
sb->s_dirt = 0;
|
2007-04-02 10:50:19 -04:00
|
|
|
}
|
|
|
|
|
2008-05-07 11:43:44 -04:00
|
|
|
static int btrfs_test_super(struct super_block *s, void *data)
|
2007-08-29 09:11:44 -04:00
|
|
|
{
|
2008-05-07 11:43:44 -04:00
|
|
|
struct btrfs_fs_devices *test_fs_devices = data;
|
|
|
|
struct btrfs_root *root = btrfs_sb(s);
|
2007-08-29 09:11:44 -04:00
|
|
|
|
2008-05-07 11:43:44 -04:00
|
|
|
return root->fs_info->fs_devices == test_fs_devices;
|
2007-08-29 09:11:44 -04:00
|
|
|
}
|
|
|
|
|
2008-06-10 10:40:29 -04:00
|
|
|
/*
|
|
|
|
* Find a superblock for the given device / mount point.
|
|
|
|
*
|
|
|
|
* Note: This is based on get_sb_bdev from fs/super.c with a few additions
|
|
|
|
* for multiple device setup. Make sure to keep it in sync.
|
|
|
|
*/
|
|
|
|
static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
|
|
|
|
const char *dev_name, void *data, struct vfsmount *mnt)
|
2007-08-29 09:11:44 -04:00
|
|
|
{
|
2008-06-10 10:40:29 -04:00
|
|
|
char *subvol_name = NULL;
|
2007-08-29 09:11:44 -04:00
|
|
|
struct block_device *bdev = NULL;
|
|
|
|
struct super_block *s;
|
|
|
|
struct dentry *root;
|
2008-03-24 15:02:07 -04:00
|
|
|
struct btrfs_fs_devices *fs_devices = NULL;
|
2008-12-02 06:36:09 -05:00
|
|
|
fmode_t mode = FMODE_READ;
|
2007-08-29 09:11:44 -04:00
|
|
|
int error = 0;
|
|
|
|
|
2008-12-02 06:36:09 -05:00
|
|
|
if (!(flags & MS_RDONLY))
|
|
|
|
mode |= FMODE_WRITE;
|
|
|
|
|
|
|
|
error = btrfs_parse_early_options(data, mode, fs_type,
|
2008-06-10 10:40:46 -04:00
|
|
|
&subvol_name, &fs_devices);
|
2008-06-10 10:40:29 -04:00
|
|
|
if (error)
|
2009-01-05 15:43:42 -05:00
|
|
|
return error;
|
2008-06-10 10:40:29 -04:00
|
|
|
|
2008-12-02 06:36:09 -05:00
|
|
|
error = btrfs_scan_one_device(dev_name, mode, fs_type, &fs_devices);
|
2008-03-24 15:02:07 -04:00
|
|
|
if (error)
|
2008-06-10 10:40:29 -04:00
|
|
|
goto error_free_subvol_name;
|
2007-08-29 09:11:44 -04:00
|
|
|
|
2008-12-02 06:36:09 -05:00
|
|
|
error = btrfs_open_devices(fs_devices, mode, fs_type);
|
2008-03-24 15:02:07 -04:00
|
|
|
if (error)
|
2008-06-10 10:40:29 -04:00
|
|
|
goto error_free_subvol_name;
|
2008-03-24 15:02:07 -04:00
|
|
|
|
2008-11-17 21:11:30 -05:00
|
|
|
if (!(flags & MS_RDONLY) && fs_devices->rw_devices == 0) {
|
|
|
|
error = -EACCES;
|
|
|
|
goto error_close_devices;
|
|
|
|
}
|
|
|
|
|
2008-05-13 13:46:40 -04:00
|
|
|
bdev = fs_devices->latest_bdev;
|
2008-05-07 11:43:44 -04:00
|
|
|
s = sget(fs_type, btrfs_test_super, set_anon_super, fs_devices);
|
2007-08-29 09:11:44 -04:00
|
|
|
if (IS_ERR(s))
|
|
|
|
goto error_s;
|
|
|
|
|
|
|
|
if (s->s_root) {
|
|
|
|
if ((flags ^ s->s_flags) & MS_RDONLY) {
|
|
|
|
up_write(&s->s_umount);
|
|
|
|
deactivate_super(s);
|
|
|
|
error = -EBUSY;
|
2008-11-12 14:34:12 -05:00
|
|
|
goto error_close_devices;
|
2007-08-29 09:11:44 -04:00
|
|
|
}
|
|
|
|
|
2008-11-17 21:11:30 -05:00
|
|
|
btrfs_close_devices(fs_devices);
|
2007-08-29 09:11:44 -04:00
|
|
|
} else {
|
|
|
|
char b[BDEVNAME_SIZE];
|
|
|
|
|
|
|
|
s->s_flags = flags;
|
|
|
|
strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id));
|
2008-03-24 15:02:07 -04:00
|
|
|
error = btrfs_fill_super(s, fs_devices, data,
|
|
|
|
flags & MS_SILENT ? 1 : 0);
|
2007-08-29 09:11:44 -04:00
|
|
|
if (error) {
|
|
|
|
up_write(&s->s_umount);
|
|
|
|
deactivate_super(s);
|
2009-01-05 15:43:42 -05:00
|
|
|
goto error_free_subvol_name;
|
2007-08-29 09:11:44 -04:00
|
|
|
}
|
|
|
|
|
2008-04-28 15:29:42 -04:00
|
|
|
btrfs_sb(s)->fs_info->bdev_holder = fs_type;
|
2007-08-29 09:11:44 -04:00
|
|
|
s->s_flags |= MS_ACTIVE;
|
|
|
|
}
|
|
|
|
|
2008-08-19 11:49:35 -04:00
|
|
|
if (!strcmp(subvol_name, "."))
|
|
|
|
root = dget(s->s_root);
|
|
|
|
else {
|
|
|
|
mutex_lock(&s->s_root->d_inode->i_mutex);
|
2009-01-05 21:25:51 -05:00
|
|
|
root = lookup_one_len(subvol_name, s->s_root,
|
|
|
|
strlen(subvol_name));
|
2008-08-19 11:49:35 -04:00
|
|
|
mutex_unlock(&s->s_root->d_inode->i_mutex);
|
2009-01-05 21:25:51 -05:00
|
|
|
|
2008-08-19 11:49:35 -04:00
|
|
|
if (IS_ERR(root)) {
|
|
|
|
up_write(&s->s_umount);
|
|
|
|
deactivate_super(s);
|
|
|
|
error = PTR_ERR(root);
|
2009-01-05 15:43:42 -05:00
|
|
|
goto error_free_subvol_name;
|
2008-08-19 11:49:35 -04:00
|
|
|
}
|
|
|
|
if (!root->d_inode) {
|
|
|
|
dput(root);
|
|
|
|
up_write(&s->s_umount);
|
|
|
|
deactivate_super(s);
|
|
|
|
error = -ENXIO;
|
2009-01-05 15:43:42 -05:00
|
|
|
goto error_free_subvol_name;
|
2008-08-19 11:49:35 -04:00
|
|
|
}
|
2007-08-29 09:11:44 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
mnt->mnt_sb = s;
|
|
|
|
mnt->mnt_root = root;
|
2008-06-10 10:40:29 -04:00
|
|
|
|
|
|
|
kfree(subvol_name);
|
2007-08-29 09:11:44 -04:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
error_s:
|
|
|
|
error = PTR_ERR(s);
|
2008-11-12 14:34:12 -05:00
|
|
|
error_close_devices:
|
2008-03-24 15:02:07 -04:00
|
|
|
btrfs_close_devices(fs_devices);
|
2008-06-10 10:40:29 -04:00
|
|
|
error_free_subvol_name:
|
|
|
|
kfree(subvol_name);
|
2007-08-29 09:11:44 -04:00
|
|
|
return error;
|
|
|
|
}
|
2007-03-21 11:12:56 -04:00
|
|
|
|
2008-11-12 14:34:12 -05:00
|
|
|
static int btrfs_remount(struct super_block *sb, int *flags, char *data)
|
|
|
|
{
|
|
|
|
struct btrfs_root *root = btrfs_sb(sb);
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (*flags & MS_RDONLY) {
|
|
|
|
sb->s_flags |= MS_RDONLY;
|
|
|
|
|
|
|
|
ret = btrfs_commit_super(root);
|
|
|
|
WARN_ON(ret);
|
|
|
|
} else {
|
2008-11-17 21:11:30 -05:00
|
|
|
if (root->fs_info->fs_devices->rw_devices == 0)
|
|
|
|
return -EACCES;
|
|
|
|
|
2008-11-12 14:34:12 -05:00
|
|
|
if (btrfs_super_log_root(&root->fs_info->super_copy) != 0)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
ret = btrfs_cleanup_reloc_trees(root);
|
|
|
|
WARN_ON(ret);
|
|
|
|
|
|
|
|
ret = btrfs_cleanup_fs_roots(root->fs_info);
|
|
|
|
WARN_ON(ret);
|
|
|
|
|
|
|
|
sb->s_flags &= ~MS_RDONLY;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-04-19 21:01:03 -04:00
|
|
|
static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
|
|
|
|
{
|
|
|
|
struct btrfs_root *root = btrfs_sb(dentry->d_sb);
|
2007-06-26 10:06:50 -04:00
|
|
|
struct btrfs_super_block *disk_super = &root->fs_info->super_copy;
|
2007-10-15 16:15:53 -04:00
|
|
|
int bits = dentry->d_sb->s_blocksize_bits;
|
2008-08-18 07:01:52 -04:00
|
|
|
__be32 *fsid = (__be32 *)root->fs_info->fsid;
|
2007-04-19 21:01:03 -04:00
|
|
|
|
|
|
|
buf->f_namelen = BTRFS_NAME_LEN;
|
2007-10-15 16:15:53 -04:00
|
|
|
buf->f_blocks = btrfs_super_total_bytes(disk_super) >> bits;
|
|
|
|
buf->f_bfree = buf->f_blocks -
|
|
|
|
(btrfs_super_bytes_used(disk_super) >> bits);
|
2007-04-19 21:01:03 -04:00
|
|
|
buf->f_bavail = buf->f_bfree;
|
|
|
|
buf->f_bsize = dentry->d_sb->s_blocksize;
|
|
|
|
buf->f_type = BTRFS_SUPER_MAGIC;
|
2009-01-05 21:25:51 -05:00
|
|
|
|
2008-08-18 07:01:52 -04:00
|
|
|
/* We treat it as constant endianness (it doesn't matter _which_)
|
2009-01-05 21:25:51 -05:00
|
|
|
because we want the fsid to come out the same whether mounted
|
2008-08-18 07:01:52 -04:00
|
|
|
on a big-endian or little-endian host */
|
|
|
|
buf->f_fsid.val[0] = be32_to_cpu(fsid[0]) ^ be32_to_cpu(fsid[2]);
|
|
|
|
buf->f_fsid.val[1] = be32_to_cpu(fsid[1]) ^ be32_to_cpu(fsid[3]);
|
2008-08-18 08:10:20 -04:00
|
|
|
/* Mask in the root object ID too, to disambiguate subvols */
|
|
|
|
buf->f_fsid.val[0] ^= BTRFS_I(dentry->d_inode)->root->objectid >> 32;
|
|
|
|
buf->f_fsid.val[1] ^= BTRFS_I(dentry->d_inode)->root->objectid;
|
|
|
|
|
2007-04-19 21:01:03 -04:00
|
|
|
return 0;
|
|
|
|
}
|
2007-04-24 11:52:22 -04:00
|
|
|
|
2007-03-21 11:12:56 -04:00
|
|
|
static struct file_system_type btrfs_fs_type = {
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.name = "btrfs",
|
|
|
|
.get_sb = btrfs_get_sb,
|
2008-05-07 11:43:44 -04:00
|
|
|
.kill_sb = kill_anon_super,
|
2007-03-21 11:12:56 -04:00
|
|
|
.fs_flags = FS_REQUIRES_DEV,
|
|
|
|
};
|
2008-03-24 15:02:04 -04:00
|
|
|
|
2008-09-29 15:18:18 -04:00
|
|
|
/*
|
|
|
|
* used by btrfsctl to scan devices when no FS is mounted
|
|
|
|
*/
|
2008-03-24 15:02:07 -04:00
|
|
|
static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
|
|
|
|
unsigned long arg)
|
|
|
|
{
|
|
|
|
struct btrfs_ioctl_vol_args *vol;
|
|
|
|
struct btrfs_fs_devices *fs_devices;
|
2008-06-09 22:17:11 -04:00
|
|
|
int ret = 0;
|
2008-03-24 15:02:07 -04:00
|
|
|
int len;
|
|
|
|
|
2009-01-05 16:57:23 -05:00
|
|
|
if (!capable(CAP_SYS_ADMIN))
|
|
|
|
return -EPERM;
|
|
|
|
|
2008-03-24 15:02:07 -04:00
|
|
|
vol = kmalloc(sizeof(*vol), GFP_KERNEL);
|
|
|
|
if (copy_from_user(vol, (void __user *)arg, sizeof(*vol))) {
|
|
|
|
ret = -EFAULT;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
len = strnlen(vol->name, BTRFS_PATH_NAME_MAX);
|
|
|
|
switch (cmd) {
|
|
|
|
case BTRFS_IOC_SCAN_DEV:
|
2008-12-02 06:36:09 -05:00
|
|
|
ret = btrfs_scan_one_device(vol->name, FMODE_READ,
|
2008-03-24 15:02:07 -04:00
|
|
|
&btrfs_fs_type, &fs_devices);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
kfree(vol);
|
2008-06-09 22:17:11 -04:00
|
|
|
return ret;
|
2008-03-24 15:02:07 -04:00
|
|
|
}
|
|
|
|
|
2009-01-10 09:09:52 -05:00
|
|
|
static int btrfs_freeze(struct super_block *sb)
|
2008-01-22 12:46:56 -05:00
|
|
|
{
|
|
|
|
struct btrfs_root *root = btrfs_sb(sb);
|
2008-06-25 16:01:31 -04:00
|
|
|
mutex_lock(&root->fs_info->transaction_kthread_mutex);
|
|
|
|
mutex_lock(&root->fs_info->cleaner_mutex);
|
2009-01-10 09:09:52 -05:00
|
|
|
return 0;
|
2008-01-22 12:46:56 -05:00
|
|
|
}
|
|
|
|
|
2009-01-10 09:09:52 -05:00
|
|
|
static int btrfs_unfreeze(struct super_block *sb)
|
2008-01-22 12:46:56 -05:00
|
|
|
{
|
|
|
|
struct btrfs_root *root = btrfs_sb(sb);
|
2008-06-25 16:01:31 -04:00
|
|
|
mutex_unlock(&root->fs_info->cleaner_mutex);
|
|
|
|
mutex_unlock(&root->fs_info->transaction_kthread_mutex);
|
2009-01-10 09:09:52 -05:00
|
|
|
return 0;
|
2008-01-22 12:46:56 -05:00
|
|
|
}
|
2007-03-21 11:12:56 -04:00
|
|
|
|
2007-03-22 12:13:20 -04:00
|
|
|
static struct super_operations btrfs_super_ops = {
|
2007-03-25 13:44:56 -04:00
|
|
|
.delete_inode = btrfs_delete_inode,
|
2007-03-22 12:13:20 -04:00
|
|
|
.put_super = btrfs_put_super,
|
2007-03-23 10:01:08 -04:00
|
|
|
.write_super = btrfs_write_super,
|
|
|
|
.sync_fs = btrfs_sync_fs,
|
2008-02-20 16:11:05 -05:00
|
|
|
.show_options = generic_show_options,
|
2007-03-26 12:00:39 -04:00
|
|
|
.write_inode = btrfs_write_inode,
|
2007-04-24 11:52:22 -04:00
|
|
|
.dirty_inode = btrfs_dirty_inode,
|
2007-04-02 10:50:19 -04:00
|
|
|
.alloc_inode = btrfs_alloc_inode,
|
|
|
|
.destroy_inode = btrfs_destroy_inode,
|
2007-04-19 21:01:03 -04:00
|
|
|
.statfs = btrfs_statfs,
|
2008-11-12 14:34:12 -05:00
|
|
|
.remount_fs = btrfs_remount,
|
2009-01-10 09:09:52 -05:00
|
|
|
.freeze_fs = btrfs_freeze,
|
|
|
|
.unfreeze_fs = btrfs_unfreeze,
|
2007-03-22 12:13:20 -04:00
|
|
|
};
|
2008-03-24 15:02:04 -04:00
|
|
|
|
|
|
|
static const struct file_operations btrfs_ctl_fops = {
|
|
|
|
.unlocked_ioctl = btrfs_control_ioctl,
|
|
|
|
.compat_ioctl = btrfs_control_ioctl,
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct miscdevice btrfs_misc = {
|
|
|
|
.minor = MISC_DYNAMIC_MINOR,
|
|
|
|
.name = "btrfs-control",
|
|
|
|
.fops = &btrfs_ctl_fops
|
|
|
|
};
|
|
|
|
|
|
|
|
static int btrfs_interface_init(void)
|
|
|
|
{
|
|
|
|
return misc_register(&btrfs_misc);
|
|
|
|
}
|
|
|
|
|
2008-12-02 09:54:17 -05:00
|
|
|
static void btrfs_interface_exit(void)
|
2008-03-24 15:02:04 -04:00
|
|
|
{
|
|
|
|
if (misc_deregister(&btrfs_misc) < 0)
|
2009-01-05 21:25:51 -05:00
|
|
|
printk(KERN_INFO "misc_deregister failed for control device");
|
2008-03-24 15:02:04 -04:00
|
|
|
}
|
|
|
|
|
2007-03-21 11:12:56 -04:00
|
|
|
static int __init init_btrfs_fs(void)
|
|
|
|
{
|
2007-04-02 10:50:19 -04:00
|
|
|
int err;
|
2007-08-29 15:47:34 -04:00
|
|
|
|
|
|
|
err = btrfs_init_sysfs();
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
|
2007-06-12 06:35:45 -04:00
|
|
|
err = btrfs_init_cachep();
|
2007-04-02 10:50:19 -04:00
|
|
|
if (err)
|
2008-06-25 16:01:31 -04:00
|
|
|
goto free_sysfs;
|
2008-01-24 16:13:08 -05:00
|
|
|
|
|
|
|
err = extent_io_init();
|
2007-11-19 10:22:33 -05:00
|
|
|
if (err)
|
|
|
|
goto free_cachep;
|
|
|
|
|
2008-01-24 16:13:08 -05:00
|
|
|
err = extent_map_init();
|
|
|
|
if (err)
|
|
|
|
goto free_extent_io;
|
|
|
|
|
2008-03-24 15:02:04 -04:00
|
|
|
err = btrfs_interface_init();
|
2007-11-19 10:22:33 -05:00
|
|
|
if (err)
|
|
|
|
goto free_extent_map;
|
Btrfs: Add zlib compression support
This is a large change for adding compression on reading and writing,
both for inline and regular extents. It does some fairly large
surgery to the writeback paths.
Compression is off by default and enabled by mount -o compress. Even
when the -o compress mount option is not used, it is possible to read
compressed extents off the disk.
If compression for a given set of pages fails to make them smaller, the
file is flagged to avoid future compression attempts later.
* While finding delalloc extents, the pages are locked before being sent down
to the delalloc handler. This allows the delalloc handler to do complex things
such as cleaning the pages, marking them writeback and starting IO on their
behalf.
* Inline extents are inserted at delalloc time now. This allows us to compress
the data before inserting the inline extent, and it allows us to insert
an inline extent that spans multiple pages.
* All of the in-memory extent representations (extent_map.c, ordered-data.c etc)
are changed to record both an in-memory size and an on disk size, as well
as a flag for compression.
From a disk format point of view, the extent pointers in the file are changed
to record the on disk size of a given extent and some encoding flags.
Space in the disk format is allocated for compression encoding, as well
as encryption and a generic 'other' field. Neither the encryption or the
'other' field are currently used.
In order to limit the amount of data read for a single random read in the
file, the size of a compressed extent is limited to 128k. This is a
software only limit, the disk format supports u64 sized compressed extents.
In order to limit the ram consumed while processing extents, the uncompressed
size of a compressed extent is limited to 256k. This is a software only limit
and will be subject to tuning later.
Checksumming is still done on compressed extents, and it is done on the
uncompressed version of the data. This way additional encodings can be
layered on without having to figure out which encoding to checksum.
Compression happens at delalloc time, which is basically singled threaded because
it is usually done by a single pdflush thread. This makes it tricky to
spread the compression load across all the cpus on the box. We'll have to
look at parallel pdflush walks of dirty inodes at a later time.
Decompression is hooked into readpages and it does spread across CPUs nicely.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-10-29 14:49:59 -04:00
|
|
|
|
2008-03-24 15:02:04 -04:00
|
|
|
err = register_filesystem(&btrfs_fs_type);
|
|
|
|
if (err)
|
|
|
|
goto unregister_ioctl;
|
2008-07-23 12:12:13 -04:00
|
|
|
|
|
|
|
printk(KERN_INFO "%s loaded\n", BTRFS_BUILD_VERSION);
|
2007-11-19 10:22:33 -05:00
|
|
|
return 0;
|
|
|
|
|
2008-03-24 15:02:04 -04:00
|
|
|
unregister_ioctl:
|
|
|
|
btrfs_interface_exit();
|
2007-11-19 10:22:33 -05:00
|
|
|
free_extent_map:
|
|
|
|
extent_map_exit();
|
2008-01-24 16:13:08 -05:00
|
|
|
free_extent_io:
|
|
|
|
extent_io_exit();
|
2007-11-19 10:22:33 -05:00
|
|
|
free_cachep:
|
|
|
|
btrfs_destroy_cachep();
|
2008-06-25 16:01:31 -04:00
|
|
|
free_sysfs:
|
2007-11-19 10:22:33 -05:00
|
|
|
btrfs_exit_sysfs();
|
|
|
|
return err;
|
2007-03-21 11:12:56 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
static void __exit exit_btrfs_fs(void)
|
|
|
|
{
|
2007-06-12 06:35:45 -04:00
|
|
|
btrfs_destroy_cachep();
|
2007-08-27 16:49:44 -04:00
|
|
|
extent_map_exit();
|
2008-01-24 16:13:08 -05:00
|
|
|
extent_io_exit();
|
2008-03-24 15:02:04 -04:00
|
|
|
btrfs_interface_exit();
|
2007-03-21 11:12:56 -04:00
|
|
|
unregister_filesystem(&btrfs_fs_type);
|
2007-08-29 15:47:34 -04:00
|
|
|
btrfs_exit_sysfs();
|
2008-03-24 15:02:07 -04:00
|
|
|
btrfs_cleanup_fs_uuids();
|
Btrfs: Add zlib compression support
This is a large change for adding compression on reading and writing,
both for inline and regular extents. It does some fairly large
surgery to the writeback paths.
Compression is off by default and enabled by mount -o compress. Even
when the -o compress mount option is not used, it is possible to read
compressed extents off the disk.
If compression for a given set of pages fails to make them smaller, the
file is flagged to avoid future compression attempts later.
* While finding delalloc extents, the pages are locked before being sent down
to the delalloc handler. This allows the delalloc handler to do complex things
such as cleaning the pages, marking them writeback and starting IO on their
behalf.
* Inline extents are inserted at delalloc time now. This allows us to compress
the data before inserting the inline extent, and it allows us to insert
an inline extent that spans multiple pages.
* All of the in-memory extent representations (extent_map.c, ordered-data.c etc)
are changed to record both an in-memory size and an on disk size, as well
as a flag for compression.
From a disk format point of view, the extent pointers in the file are changed
to record the on disk size of a given extent and some encoding flags.
Space in the disk format is allocated for compression encoding, as well
as encryption and a generic 'other' field. Neither the encryption or the
'other' field are currently used.
In order to limit the amount of data read for a single random read in the
file, the size of a compressed extent is limited to 128k. This is a
software only limit, the disk format supports u64 sized compressed extents.
In order to limit the ram consumed while processing extents, the uncompressed
size of a compressed extent is limited to 256k. This is a software only limit
and will be subject to tuning later.
Checksumming is still done on compressed extents, and it is done on the
uncompressed version of the data. This way additional encodings can be
layered on without having to figure out which encoding to checksum.
Compression happens at delalloc time, which is basically singled threaded because
it is usually done by a single pdflush thread. This makes it tricky to
spread the compression load across all the cpus on the box. We'll have to
look at parallel pdflush walks of dirty inodes at a later time.
Decompression is hooked into readpages and it does spread across CPUs nicely.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-10-29 14:49:59 -04:00
|
|
|
btrfs_zlib_exit();
|
2007-03-21 11:12:56 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
module_init(init_btrfs_fs)
|
|
|
|
module_exit(exit_btrfs_fs)
|
|
|
|
|
|
|
|
MODULE_LICENSE("GPL");
|