linux/fs/btrfs/file.c

1233 lines
33 KiB
C
Raw Normal View History

/*
* 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.
*/
#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/backing-dev.h>
#include <linux/mpage.h>
#include <linux/swap.h>
#include <linux/writeback.h>
#include <linux/statfs.h>
#include <linux/compat.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
#include "ioctl.h"
#include "print-tree.h"
#include "tree-log.h"
#include "locking.h"
Btrfs: Add workaround for AppArmor changing remove_suid() In openSUSE 10.3, AppArmor modifies remove_suid to take a struct path rather than just a dentry. This patch tests that the kernel is openSUSE 10.3 or newer and adjusts the call accordingly. Debian/Ubuntu with AppArmor applied will also need a similar patch. Maintainers of btrfs under those distributions should build on this patch or, alternatively, alter their package descriptions to add -DREMOVE_SUID_PATH to the compiler command line. Signed-off-by: Jeff Mahoney <jeffm@suse.com> - --- /dev/null 1970-01-01 00:00:00.000000000 +0000 +++ b/compat.h 2008-02-06 16:46:13.000000000 -0500 @@ -0,0 +1,15 @@ +#ifndef _COMPAT_H_ +#define _COMPAT_H_ + + +/* + * Even if AppArmor isn't enabled, it still has different prototypes. + * Add more distro/version pairs here to declare which has AppArmor applied. + */ +#if defined(CONFIG_SUSE_KERNEL) +# if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) +# define REMOVE_SUID_PATH 1 +# endif +#endif + +#endif /* _COMPAT_H_ */ - --- a/file.c 2008-02-06 11:37:39.000000000 -0500 +++ b/file.c 2008-02-06 16:46:23.000000000 -0500 @@ -37,6 +37,7 @@ #include "ordered-data.h" #include "ioctl.h" #include "print-tree.h" +#include "compat.h" static int btrfs_copy_from_user(loff_t pos, int num_pages, int write_bytes, @@ -790,7 +791,11 @@ static ssize_t btrfs_file_write(struct f goto out_nolock; if (count == 0) goto out_nolock; +#ifdef REMOVE_SUID_PATH + err = remove_suid(&file->f_path); +#else err = remove_suid(fdentry(file)); +#endif if (err) goto out_nolock; file_update_time(file); Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-05-02 23:03:58 +04:00
#include "compat.h"
/* simple helper to fault in pages and copy. This should go away
* and be replaced with calls into generic code.
*/
static noinline int btrfs_copy_from_user(loff_t pos, int num_pages,
int write_bytes,
struct page **prepared_pages,
const char __user *buf)
{
long page_fault = 0;
int i;
int offset = pos & (PAGE_CACHE_SIZE - 1);
for (i = 0; i < num_pages && write_bytes > 0; i++, offset = 0) {
size_t count = min_t(size_t,
PAGE_CACHE_SIZE - offset, write_bytes);
struct page *page = prepared_pages[i];
fault_in_pages_readable(buf, count);
/* Copy data from userspace to the current page */
kmap(page);
page_fault = __copy_from_user(page_address(page) + offset,
buf, count);
/* Flush processor's dcache for this page */
flush_dcache_page(page);
kunmap(page);
buf += count;
write_bytes -= count;
if (page_fault)
break;
}
return page_fault ? -EFAULT : 0;
}
/*
* unlocks pages after btrfs_file_write is done with them
*/
static noinline void btrfs_drop_pages(struct page **pages, size_t num_pages)
{
size_t i;
for (i = 0; i < num_pages; i++) {
if (!pages[i])
break;
/* page checked is some magic around finding pages that
* have been modified without going through btrfs_set_page_dirty
* clear it here
*/
ClearPageChecked(pages[i]);
unlock_page(pages[i]);
mark_page_accessed(pages[i]);
page_cache_release(pages[i]);
}
}
/*
* after copy_from_user, pages need to be dirtied and we need to make
* sure holes are created between the current EOF and the start of
* any next extents (if required).
*
* this also makes the decision about creating an inline extent vs
* doing real data extents, marking pages dirty and delalloc as required.
*/
static noinline int dirty_and_release_pages(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct file *file,
struct page **pages,
size_t num_pages,
loff_t pos,
size_t write_bytes)
{
int err = 0;
int i;
struct inode *inode = fdentry(file)->d_inode;
u64 num_bytes;
u64 start_pos;
u64 end_of_last_block;
u64 end_pos = pos + write_bytes;
loff_t isize = i_size_read(inode);
start_pos = pos & ~((u64)root->sectorsize - 1);
num_bytes = (write_bytes + pos - start_pos +
root->sectorsize - 1) & ~((u64)root->sectorsize - 1);
end_of_last_block = start_pos + num_bytes - 1;
Btrfs: proper -ENOSPC handling At the start of a transaction we do a btrfs_reserve_metadata_space() and specify how many items we plan on modifying. Then once we've done our modifications and such, just call btrfs_unreserve_metadata_space() for the same number of items we reserved. For keeping track of metadata needed for data I've had to add an extent_io op for when we merge extents. This lets us track space properly when we are doing sequential writes, so we don't end up reserving way more metadata space than what we need. The only place where the metadata space accounting is not done is in the relocation code. This is because Yan is going to be reworking that code in the near future, so running btrfs-vol -b could still possibly result in a ENOSPC related panic. This patch also turns off the metadata_ratio stuff in order to allow users to more efficiently use their disk space. This patch makes it so we track how much metadata we need for an inode's delayed allocation extents by tracking how many extents are currently waiting for allocation. It introduces two new callbacks for the extent_io tree's, merge_extent_hook and split_extent_hook. These help us keep track of when we merge delalloc extents together and split them up. Reservations are handled prior to any actually dirty'ing occurs, and then we unreserve after we dirty. btrfs_unreserve_metadata_for_delalloc() will make the appropriate unreservations as needed based on the number of reservations we currently have and the number of extents we currently have. Doing the reservation outside of doing any of the actual dirty'ing lets us do things like filemap_flush() the inode to try and force delalloc to happen, or as a last resort actually start allocation on all delalloc inodes in the fs. This has survived dbench, fs_mark and an fsx torture test. Signed-off-by: Josef Bacik <jbacik@redhat.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-09-12 00:12:44 +04:00
err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block);
if (err)
return err;
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 21:49:59 +03:00
for (i = 0; i < num_pages; i++) {
struct page *p = pages[i];
SetPageUptodate(p);
ClearPageChecked(p);
set_page_dirty(p);
}
if (end_pos > isize) {
i_size_write(inode, end_pos);
/* we've only changed i_size in ram, and we haven't updated
* the disk i_size. There is no need to log the inode
* at this time.
*/
}
return err;
}
/*
* this drops all the extents in the cache that intersect the range
* [start, end]. Existing extents are split as required.
*/
int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
int skip_pinned)
{
struct extent_map *em;
struct extent_map *split = NULL;
struct extent_map *split2 = NULL;
struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
u64 len = end - start + 1;
int ret;
int testend = 1;
unsigned long flags;
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 21:49:59 +03:00
int compressed = 0;
WARN_ON(end < start);
if (end == (u64)-1) {
len = (u64)-1;
testend = 0;
}
while (1) {
if (!split)
split = alloc_extent_map(GFP_NOFS);
if (!split2)
split2 = alloc_extent_map(GFP_NOFS);
write_lock(&em_tree->lock);
em = lookup_extent_mapping(em_tree, start, len);
if (!em) {
write_unlock(&em_tree->lock);
break;
}
flags = em->flags;
if (skip_pinned && test_bit(EXTENT_FLAG_PINNED, &em->flags)) {
if (em->start <= start &&
(!testend || em->start + em->len >= start + len)) {
free_extent_map(em);
write_unlock(&em_tree->lock);
break;
}
if (start < em->start) {
len = em->start - start;
} else {
len = start + len - (em->start + em->len);
start = em->start + em->len;
}
free_extent_map(em);
write_unlock(&em_tree->lock);
continue;
}
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 21:49:59 +03:00
compressed = test_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
clear_bit(EXTENT_FLAG_PINNED, &em->flags);
remove_extent_mapping(em_tree, em);
if (em->block_start < EXTENT_MAP_LAST_BYTE &&
em->start < start) {
split->start = em->start;
split->len = start - em->start;
split->orig_start = em->orig_start;
split->block_start = em->block_start;
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 21:49:59 +03:00
if (compressed)
split->block_len = em->block_len;
else
split->block_len = split->len;
split->bdev = em->bdev;
split->flags = flags;
ret = add_extent_mapping(em_tree, split);
BUG_ON(ret);
free_extent_map(split);
split = split2;
split2 = NULL;
}
if (em->block_start < EXTENT_MAP_LAST_BYTE &&
testend && em->start + em->len > start + len) {
u64 diff = start + len - em->start;
split->start = start + len;
split->len = em->start + em->len - (start + len);
split->bdev = em->bdev;
split->flags = flags;
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 21:49:59 +03:00
if (compressed) {
split->block_len = em->block_len;
split->block_start = em->block_start;
split->orig_start = em->orig_start;
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 21:49:59 +03:00
} else {
split->block_len = split->len;
split->block_start = em->block_start + diff;
split->orig_start = split->start;
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 21:49:59 +03:00
}
ret = add_extent_mapping(em_tree, split);
BUG_ON(ret);
free_extent_map(split);
split = NULL;
}
write_unlock(&em_tree->lock);
/* once for us */
free_extent_map(em);
/* once for the tree*/
free_extent_map(em);
}
if (split)
free_extent_map(split);
if (split2)
free_extent_map(split2);
return 0;
}
/*
* this is very complex, but the basic idea is to drop all extents
* in the range start - end. hint_block is filled in with a block number
* that would be a good hint to the block allocator for this file.
*
* If an extent intersects the range but is not entirely inside the range
* it is either truncated or split. Anything entirely inside the range
* is deleted from the tree.
*
* inline_limit is used to tell this code which offsets in the file to keep
* if they contain inline extents.
*/
noinline int btrfs_drop_extents(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct inode *inode,
u64 start, u64 end, u64 locked_end,
u64 inline_limit, u64 *hint_byte, int drop_cache)
{
u64 extent_end = 0;
u64 search_start = start;
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 21:49:59 +03:00
u64 ram_bytes = 0;
u64 disk_bytenr = 0;
u64 orig_locked_end = locked_end;
u8 compression;
u8 encryption;
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 21:49:59 +03:00
u16 other_encoding = 0;
struct extent_buffer *leaf;
struct btrfs_file_extent_item *extent;
struct btrfs_path *path;
struct btrfs_key key;
struct btrfs_file_extent_item old;
int keep;
int slot;
int bookend;
int found_type = 0;
int found_extent;
int found_inline;
int recow;
int ret;
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 21:49:59 +03:00
inline_limit = 0;
if (drop_cache)
btrfs_drop_extent_cache(inode, start, end - 1, 0);
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
while (1) {
recow = 0;
btrfs_release_path(root, path);
ret = btrfs_lookup_file_extent(trans, root, path, inode->i_ino,
search_start, -1);
if (ret < 0)
goto out;
if (ret > 0) {
if (path->slots[0] == 0) {
ret = 0;
goto out;
}
path->slots[0]--;
}
next_slot:
keep = 0;
bookend = 0;
found_extent = 0;
found_inline = 0;
compression = 0;
encryption = 0;
extent = NULL;
leaf = path->nodes[0];
slot = path->slots[0];
ret = 0;
btrfs_item_key_to_cpu(leaf, &key, slot);
if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY &&
key.offset >= end) {
goto out;
}
if (btrfs_key_type(&key) > BTRFS_EXTENT_DATA_KEY ||
key.objectid != inode->i_ino) {
goto out;
}
if (recow) {
search_start = max(key.offset, start);
continue;
}
if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY) {
extent = btrfs_item_ptr(leaf, slot,
struct btrfs_file_extent_item);
found_type = btrfs_file_extent_type(leaf, extent);
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 21:49:59 +03:00
compression = btrfs_file_extent_compression(leaf,
extent);
encryption = btrfs_file_extent_encryption(leaf,
extent);
other_encoding = btrfs_file_extent_other_encoding(leaf,
extent);
if (found_type == BTRFS_FILE_EXTENT_REG ||
found_type == BTRFS_FILE_EXTENT_PREALLOC) {
extent_end =
btrfs_file_extent_disk_bytenr(leaf,
extent);
if (extent_end)
*hint_byte = extent_end;
extent_end = key.offset +
btrfs_file_extent_num_bytes(leaf, extent);
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 21:49:59 +03:00
ram_bytes = btrfs_file_extent_ram_bytes(leaf,
extent);
found_extent = 1;
} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
found_inline = 1;
extent_end = key.offset +
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 21:49:59 +03:00
btrfs_file_extent_inline_len(leaf, extent);
}
} else {
extent_end = search_start;
}
/* we found nothing we can drop */
if ((!found_extent && !found_inline) ||
search_start >= extent_end) {
int nextret;
u32 nritems;
nritems = btrfs_header_nritems(leaf);
if (slot >= nritems - 1) {
nextret = btrfs_next_leaf(root, path);
if (nextret)
goto out;
recow = 1;
} else {
path->slots[0]++;
}
goto next_slot;
}
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 21:49:59 +03:00
if (end <= extent_end && start >= key.offset && found_inline)
*hint_byte = EXTENT_MAP_INLINE;
if (found_extent) {
read_extent_buffer(leaf, &old, (unsigned long)extent,
sizeof(old));
}
if (end < extent_end && end >= key.offset) {
bookend = 1;
if (found_inline && start <= key.offset)
keep = 1;
}
if (bookend && found_extent) {
if (locked_end < extent_end) {
ret = try_lock_extent(&BTRFS_I(inode)->io_tree,
locked_end, extent_end - 1,
GFP_NOFS);
if (!ret) {
btrfs_release_path(root, path);
lock_extent(&BTRFS_I(inode)->io_tree,
locked_end, extent_end - 1,
GFP_NOFS);
locked_end = extent_end;
continue;
}
locked_end = extent_end;
}
disk_bytenr = le64_to_cpu(old.disk_bytenr);
if (disk_bytenr != 0) {
ret = btrfs_inc_extent_ref(trans, root,
disk_bytenr,
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE) This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-06-10 18:45:14 +04:00
le64_to_cpu(old.disk_num_bytes), 0,
root->root_key.objectid,
key.objectid, key.offset -
le64_to_cpu(old.offset));
BUG_ON(ret);
}
}
if (found_inline) {
u64 mask = root->sectorsize - 1;
search_start = (extent_end + mask) & ~mask;
} else
search_start = extent_end;
/* truncate existing extent */
if (start > key.offset) {
u64 new_num;
u64 old_num;
keep = 1;
WARN_ON(start & (root->sectorsize - 1));
if (found_extent) {
new_num = start - key.offset;
old_num = btrfs_file_extent_num_bytes(leaf,
extent);
*hint_byte =
btrfs_file_extent_disk_bytenr(leaf,
extent);
if (btrfs_file_extent_disk_bytenr(leaf,
extent)) {
inode_sub_bytes(inode, old_num -
new_num);
}
btrfs_set_file_extent_num_bytes(leaf,
extent, new_num);
btrfs_mark_buffer_dirty(leaf);
} else if (key.offset < inline_limit &&
(end > extent_end) &&
(inline_limit < extent_end)) {
u32 new_size;
new_size = btrfs_file_extent_calc_inline_size(
inline_limit - key.offset);
inode_sub_bytes(inode, extent_end -
inline_limit);
btrfs_set_file_extent_ram_bytes(leaf, extent,
new_size);
if (!compression && !encryption) {
btrfs_truncate_item(trans, root, path,
new_size, 1);
}
}
}
/* delete the entire extent */
if (!keep) {
if (found_inline)
inode_sub_bytes(inode, extent_end -
key.offset);
ret = btrfs_del_item(trans, root, path);
/* TODO update progress marker and return */
BUG_ON(ret);
extent = NULL;
btrfs_release_path(root, path);
/* the extent will be freed later */
}
if (bookend && found_inline && start <= key.offset) {
u32 new_size;
new_size = btrfs_file_extent_calc_inline_size(
extent_end - end);
inode_sub_bytes(inode, end - key.offset);
btrfs_set_file_extent_ram_bytes(leaf, extent,
new_size);
if (!compression && !encryption)
ret = btrfs_truncate_item(trans, root, path,
new_size, 0);
BUG_ON(ret);
}
/* create bookend, splitting the extent in two */
if (bookend && found_extent) {
struct btrfs_key ins;
ins.objectid = inode->i_ino;
ins.offset = end;
btrfs_set_key_type(&ins, BTRFS_EXTENT_DATA_KEY);
btrfs_release_path(root, path);
path->leave_spinning = 1;
ret = btrfs_insert_empty_item(trans, root, path, &ins,
sizeof(*extent));
BUG_ON(ret);
leaf = path->nodes[0];
extent = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
write_extent_buffer(leaf, &old,
(unsigned long)extent, sizeof(old));
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 21:49:59 +03:00
btrfs_set_file_extent_compression(leaf, extent,
compression);
btrfs_set_file_extent_encryption(leaf, extent,
encryption);
btrfs_set_file_extent_other_encoding(leaf, extent,
other_encoding);
btrfs_set_file_extent_offset(leaf, extent,
le64_to_cpu(old.offset) + end - key.offset);
WARN_ON(le64_to_cpu(old.num_bytes) <
(extent_end - end));
btrfs_set_file_extent_num_bytes(leaf, extent,
extent_end - end);
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 21:49:59 +03:00
/*
* set the ram bytes to the size of the full extent
* before splitting. This is a worst case flag,
* but its the best we can do because we don't know
* how splitting affects compression
*/
btrfs_set_file_extent_ram_bytes(leaf, extent,
ram_bytes);
btrfs_set_file_extent_type(leaf, extent, found_type);
btrfs_unlock_up_safe(path, 1);
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_set_lock_blocking(path->nodes[0]);
path->leave_spinning = 0;
btrfs_release_path(root, path);
if (disk_bytenr != 0)
inode_add_bytes(inode, extent_end - end);
}
if (found_extent && !keep) {
u64 old_disk_bytenr = le64_to_cpu(old.disk_bytenr);
if (old_disk_bytenr != 0) {
inode_sub_bytes(inode,
le64_to_cpu(old.num_bytes));
ret = btrfs_free_extent(trans, root,
old_disk_bytenr,
le64_to_cpu(old.disk_num_bytes),
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE) This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-06-10 18:45:14 +04:00
0, root->root_key.objectid,
key.objectid, key.offset -
le64_to_cpu(old.offset));
BUG_ON(ret);
*hint_byte = old_disk_bytenr;
}
}
if (search_start >= end) {
ret = 0;
goto out;
}
}
out:
btrfs_free_path(path);
if (locked_end > orig_locked_end) {
unlock_extent(&BTRFS_I(inode)->io_tree, orig_locked_end,
locked_end - 1, GFP_NOFS);
}
return ret;
}
static int extent_mergeable(struct extent_buffer *leaf, int slot,
u64 objectid, u64 bytenr, u64 *start, u64 *end)
{
struct btrfs_file_extent_item *fi;
struct btrfs_key key;
u64 extent_end;
if (slot < 0 || slot >= btrfs_header_nritems(leaf))
return 0;
btrfs_item_key_to_cpu(leaf, &key, slot);
if (key.objectid != objectid || key.type != BTRFS_EXTENT_DATA_KEY)
return 0;
fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG ||
btrfs_file_extent_disk_bytenr(leaf, fi) != bytenr ||
btrfs_file_extent_compression(leaf, fi) ||
btrfs_file_extent_encryption(leaf, fi) ||
btrfs_file_extent_other_encoding(leaf, fi))
return 0;
extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi);
if ((*start && *start != key.offset) || (*end && *end != extent_end))
return 0;
*start = key.offset;
*end = extent_end;
return 1;
}
/*
* Mark extent in the range start - end as written.
*
* This changes extent type from 'pre-allocated' to 'regular'. If only
* part of extent is marked as written, the extent will be split into
* two or three.
*/
int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct inode *inode, u64 start, u64 end)
{
struct extent_buffer *leaf;
struct btrfs_path *path;
struct btrfs_file_extent_item *fi;
struct btrfs_key key;
u64 bytenr;
u64 num_bytes;
u64 extent_end;
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE) This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-06-10 18:45:14 +04:00
u64 orig_offset;
u64 other_start;
u64 other_end;
u64 split = start;
u64 locked_end = end;
int extent_type;
int split_end = 1;
int ret;
btrfs_drop_extent_cache(inode, start, end - 1, 0);
path = btrfs_alloc_path();
BUG_ON(!path);
again:
key.objectid = inode->i_ino;
key.type = BTRFS_EXTENT_DATA_KEY;
if (split == start)
key.offset = split;
else
key.offset = split - 1;
ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
if (ret > 0 && path->slots[0] > 0)
path->slots[0]--;
leaf = path->nodes[0];
btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
BUG_ON(key.objectid != inode->i_ino ||
key.type != BTRFS_EXTENT_DATA_KEY);
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
extent_type = btrfs_file_extent_type(leaf, fi);
BUG_ON(extent_type != BTRFS_FILE_EXTENT_PREALLOC);
extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi);
BUG_ON(key.offset > start || extent_end < end);
bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE) This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-06-10 18:45:14 +04:00
orig_offset = key.offset - btrfs_file_extent_offset(leaf, fi);
if (key.offset == start)
split = end;
if (key.offset == start && extent_end == end) {
int del_nr = 0;
int del_slot = 0;
other_start = end;
other_end = 0;
if (extent_mergeable(leaf, path->slots[0] + 1, inode->i_ino,
bytenr, &other_start, &other_end)) {
extent_end = other_end;
del_slot = path->slots[0] + 1;
del_nr++;
ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE) This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-06-10 18:45:14 +04:00
0, root->root_key.objectid,
inode->i_ino, orig_offset);
BUG_ON(ret);
}
other_start = 0;
other_end = start;
if (extent_mergeable(leaf, path->slots[0] - 1, inode->i_ino,
bytenr, &other_start, &other_end)) {
key.offset = other_start;
del_slot = path->slots[0];
del_nr++;
ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE) This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-06-10 18:45:14 +04:00
0, root->root_key.objectid,
inode->i_ino, orig_offset);
BUG_ON(ret);
}
split_end = 0;
if (del_nr == 0) {
btrfs_set_file_extent_type(leaf, fi,
BTRFS_FILE_EXTENT_REG);
goto done;
}
fi = btrfs_item_ptr(leaf, del_slot - 1,
struct btrfs_file_extent_item);
btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG);
btrfs_set_file_extent_num_bytes(leaf, fi,
extent_end - key.offset);
btrfs_mark_buffer_dirty(leaf);
ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
BUG_ON(ret);
goto release;
} else if (split == start) {
if (locked_end < extent_end) {
ret = try_lock_extent(&BTRFS_I(inode)->io_tree,
locked_end, extent_end - 1, GFP_NOFS);
if (!ret) {
btrfs_release_path(root, path);
lock_extent(&BTRFS_I(inode)->io_tree,
locked_end, extent_end - 1, GFP_NOFS);
locked_end = extent_end;
goto again;
}
locked_end = extent_end;
}
btrfs_set_file_extent_num_bytes(leaf, fi, split - key.offset);
} else {
BUG_ON(key.offset != start);
key.offset = split;
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE) This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-06-10 18:45:14 +04:00
btrfs_set_file_extent_offset(leaf, fi, key.offset -
orig_offset);
btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - split);
btrfs_set_item_key_safe(trans, root, path, &key);
extent_end = split;
}
if (extent_end == end) {
split_end = 0;
extent_type = BTRFS_FILE_EXTENT_REG;
}
if (extent_end == end && split == start) {
other_start = end;
other_end = 0;
if (extent_mergeable(leaf, path->slots[0] + 1, inode->i_ino,
bytenr, &other_start, &other_end)) {
path->slots[0]++;
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
key.offset = split;
btrfs_set_item_key_safe(trans, root, path, &key);
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE) This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-06-10 18:45:14 +04:00
btrfs_set_file_extent_offset(leaf, fi, key.offset -
orig_offset);
btrfs_set_file_extent_num_bytes(leaf, fi,
other_end - split);
goto done;
}
}
if (extent_end == end && split == end) {
other_start = 0;
other_end = start;
if (extent_mergeable(leaf, path->slots[0] - 1 , inode->i_ino,
bytenr, &other_start, &other_end)) {
path->slots[0]--;
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
btrfs_set_file_extent_num_bytes(leaf, fi, extent_end -
other_start);
goto done;
}
}
btrfs_mark_buffer_dirty(leaf);
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE) This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-06-10 18:45:14 +04:00
ret = btrfs_inc_extent_ref(trans, root, bytenr, num_bytes, 0,
root->root_key.objectid,
inode->i_ino, orig_offset);
BUG_ON(ret);
btrfs_release_path(root, path);
key.offset = start;
ret = btrfs_insert_empty_item(trans, root, path, &key, sizeof(*fi));
BUG_ON(ret);
leaf = path->nodes[0];
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
btrfs_set_file_extent_generation(leaf, fi, trans->transid);
btrfs_set_file_extent_type(leaf, fi, extent_type);
btrfs_set_file_extent_disk_bytenr(leaf, fi, bytenr);
btrfs_set_file_extent_disk_num_bytes(leaf, fi, num_bytes);
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE) This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-06-10 18:45:14 +04:00
btrfs_set_file_extent_offset(leaf, fi, key.offset - orig_offset);
btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - key.offset);
btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes);
btrfs_set_file_extent_compression(leaf, fi, 0);
btrfs_set_file_extent_encryption(leaf, fi, 0);
btrfs_set_file_extent_other_encoding(leaf, fi, 0);
done:
btrfs_mark_buffer_dirty(leaf);
release:
btrfs_release_path(root, path);
if (split_end && split == start) {
split = end;
goto again;
}
if (locked_end > end) {
unlock_extent(&BTRFS_I(inode)->io_tree, end, locked_end - 1,
GFP_NOFS);
}
btrfs_free_path(path);
return 0;
}
/*
* this gets pages into the page cache and locks them down, it also properly
* waits for data=ordered extents to finish before allowing the pages to be
* modified.
*/
static noinline int prepare_pages(struct btrfs_root *root, struct file *file,
struct page **pages, size_t num_pages,
loff_t pos, unsigned long first_index,
unsigned long last_index, size_t write_bytes)
{
int i;
unsigned long index = pos >> PAGE_CACHE_SHIFT;
struct inode *inode = fdentry(file)->d_inode;
int err = 0;
u64 start_pos;
u64 last_pos;
start_pos = pos & ~((u64)root->sectorsize - 1);
last_pos = ((u64)index + num_pages) << PAGE_CACHE_SHIFT;
if (start_pos > inode->i_size) {
err = btrfs_cont_expand(inode, start_pos);
if (err)
return err;
}
memset(pages, 0, num_pages * sizeof(struct page *));
again:
for (i = 0; i < num_pages; i++) {
pages[i] = grab_cache_page(inode->i_mapping, index + i);
if (!pages[i]) {
err = -ENOMEM;
BUG_ON(1);
}
wait_on_page_writeback(pages[i]);
}
if (start_pos < inode->i_size) {
struct btrfs_ordered_extent *ordered;
lock_extent(&BTRFS_I(inode)->io_tree,
start_pos, last_pos - 1, GFP_NOFS);
ordered = btrfs_lookup_first_ordered_extent(inode,
last_pos - 1);
if (ordered &&
ordered->file_offset + ordered->len > start_pos &&
ordered->file_offset < last_pos) {
btrfs_put_ordered_extent(ordered);
unlock_extent(&BTRFS_I(inode)->io_tree,
start_pos, last_pos - 1, GFP_NOFS);
for (i = 0; i < num_pages; i++) {
unlock_page(pages[i]);
page_cache_release(pages[i]);
}
btrfs_wait_ordered_range(inode, start_pos,
last_pos - start_pos);
goto again;
}
if (ordered)
btrfs_put_ordered_extent(ordered);
clear_extent_bits(&BTRFS_I(inode)->io_tree, start_pos,
last_pos - 1, EXTENT_DIRTY | EXTENT_DELALLOC |
EXTENT_DO_ACCOUNTING,
GFP_NOFS);
unlock_extent(&BTRFS_I(inode)->io_tree,
start_pos, last_pos - 1, GFP_NOFS);
}
for (i = 0; i < num_pages; i++) {
clear_page_dirty_for_io(pages[i]);
set_page_extent_mapped(pages[i]);
WARN_ON(!PageLocked(pages[i]));
}
return 0;
}
static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
size_t count, loff_t *ppos)
{
loff_t pos;
loff_t start_pos;
ssize_t num_written = 0;
ssize_t err = 0;
int ret = 0;
struct inode *inode = fdentry(file)->d_inode;
struct btrfs_root *root = BTRFS_I(inode)->root;
struct page **pages = NULL;
int nrptrs;
struct page *pinned[2];
unsigned long first_index;
unsigned long last_index;
int will_write;
will_write = ((file->f_flags & O_SYNC) || IS_SYNC(inode) ||
(file->f_flags & O_DIRECT));
nrptrs = min((count + PAGE_CACHE_SIZE - 1) / PAGE_CACHE_SIZE,
PAGE_CACHE_SIZE / (sizeof(struct page *)));
pinned[0] = NULL;
pinned[1] = NULL;
pos = *ppos;
start_pos = pos;
vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
/* do the reserve before the mutex lock in case we have to do some
* flushing. We wouldn't deadlock, but this is more polite.
*/
err = btrfs_reserve_metadata_for_delalloc(root, inode, 1);
if (err)
goto out_nolock;
mutex_lock(&inode->i_mutex);
current->backing_dev_info = inode->i_mapping->backing_dev_info;
err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode));
if (err)
goto out;
if (count == 0)
goto out;
err = file_remove_suid(file);
if (err)
goto out;
Btrfs: proper -ENOSPC handling At the start of a transaction we do a btrfs_reserve_metadata_space() and specify how many items we plan on modifying. Then once we've done our modifications and such, just call btrfs_unreserve_metadata_space() for the same number of items we reserved. For keeping track of metadata needed for data I've had to add an extent_io op for when we merge extents. This lets us track space properly when we are doing sequential writes, so we don't end up reserving way more metadata space than what we need. The only place where the metadata space accounting is not done is in the relocation code. This is because Yan is going to be reworking that code in the near future, so running btrfs-vol -b could still possibly result in a ENOSPC related panic. This patch also turns off the metadata_ratio stuff in order to allow users to more efficiently use their disk space. This patch makes it so we track how much metadata we need for an inode's delayed allocation extents by tracking how many extents are currently waiting for allocation. It introduces two new callbacks for the extent_io tree's, merge_extent_hook and split_extent_hook. These help us keep track of when we merge delalloc extents together and split them up. Reservations are handled prior to any actually dirty'ing occurs, and then we unreserve after we dirty. btrfs_unreserve_metadata_for_delalloc() will make the appropriate unreservations as needed based on the number of reservations we currently have and the number of extents we currently have. Doing the reservation outside of doing any of the actual dirty'ing lets us do things like filemap_flush() the inode to try and force delalloc to happen, or as a last resort actually start allocation on all delalloc inodes in the fs. This has survived dbench, fs_mark and an fsx torture test. Signed-off-by: Josef Bacik <jbacik@redhat.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-09-12 00:12:44 +04:00
file_update_time(file);
pages = kmalloc(nrptrs * sizeof(struct page *), GFP_KERNEL);
/* generic_write_checks can change our pos */
start_pos = pos;
BTRFS_I(inode)->sequence++;
first_index = pos >> PAGE_CACHE_SHIFT;
last_index = (pos + count) >> PAGE_CACHE_SHIFT;
/*
* there are lots of better ways to do this, but this code
* makes sure the first and last page in the file range are
* up to date and ready for cow
*/
if ((pos & (PAGE_CACHE_SIZE - 1))) {
pinned[0] = grab_cache_page(inode->i_mapping, first_index);
if (!PageUptodate(pinned[0])) {
ret = btrfs_readpage(NULL, pinned[0]);
BUG_ON(ret);
wait_on_page_locked(pinned[0]);
} else {
unlock_page(pinned[0]);
}
}
if ((pos + count) & (PAGE_CACHE_SIZE - 1)) {
pinned[1] = grab_cache_page(inode->i_mapping, last_index);
if (!PageUptodate(pinned[1])) {
ret = btrfs_readpage(NULL, pinned[1]);
BUG_ON(ret);
wait_on_page_locked(pinned[1]);
} else {
unlock_page(pinned[1]);
}
}
while (count > 0) {
size_t offset = pos & (PAGE_CACHE_SIZE - 1);
size_t write_bytes = min(count, nrptrs *
(size_t)PAGE_CACHE_SIZE -
offset);
size_t num_pages = (write_bytes + PAGE_CACHE_SIZE - 1) >>
PAGE_CACHE_SHIFT;
WARN_ON(num_pages > nrptrs);
memset(pages, 0, sizeof(struct page *) * nrptrs);
ret = btrfs_check_data_free_space(root, inode, write_bytes);
if (ret)
goto out;
ret = prepare_pages(root, file, pages, num_pages,
pos, first_index, last_index,
write_bytes);
if (ret) {
btrfs_free_reserved_data_space(root, inode,
write_bytes);
goto out;
}
ret = btrfs_copy_from_user(pos, num_pages,
write_bytes, pages, buf);
if (ret) {
btrfs_free_reserved_data_space(root, inode,
write_bytes);
btrfs_drop_pages(pages, num_pages);
goto out;
}
ret = dirty_and_release_pages(NULL, root, file, pages,
num_pages, pos, write_bytes);
btrfs_drop_pages(pages, num_pages);
if (ret) {
btrfs_free_reserved_data_space(root, inode,
write_bytes);
goto out;
}
if (will_write) {
btrfs_fdatawrite_range(inode->i_mapping, pos,
pos + write_bytes - 1,
WB_SYNC_ALL);
} else {
balance_dirty_pages_ratelimited_nr(inode->i_mapping,
num_pages);
if (num_pages <
(root->leafsize >> PAGE_CACHE_SHIFT) + 1)
btrfs_btree_balance_dirty(root, 1);
btrfs_throttle(root);
}
buf += write_bytes;
count -= write_bytes;
pos += write_bytes;
num_written += write_bytes;
cond_resched();
}
out:
mutex_unlock(&inode->i_mutex);
if (ret)
err = ret;
Btrfs: proper -ENOSPC handling At the start of a transaction we do a btrfs_reserve_metadata_space() and specify how many items we plan on modifying. Then once we've done our modifications and such, just call btrfs_unreserve_metadata_space() for the same number of items we reserved. For keeping track of metadata needed for data I've had to add an extent_io op for when we merge extents. This lets us track space properly when we are doing sequential writes, so we don't end up reserving way more metadata space than what we need. The only place where the metadata space accounting is not done is in the relocation code. This is because Yan is going to be reworking that code in the near future, so running btrfs-vol -b could still possibly result in a ENOSPC related panic. This patch also turns off the metadata_ratio stuff in order to allow users to more efficiently use their disk space. This patch makes it so we track how much metadata we need for an inode's delayed allocation extents by tracking how many extents are currently waiting for allocation. It introduces two new callbacks for the extent_io tree's, merge_extent_hook and split_extent_hook. These help us keep track of when we merge delalloc extents together and split them up. Reservations are handled prior to any actually dirty'ing occurs, and then we unreserve after we dirty. btrfs_unreserve_metadata_for_delalloc() will make the appropriate unreservations as needed based on the number of reservations we currently have and the number of extents we currently have. Doing the reservation outside of doing any of the actual dirty'ing lets us do things like filemap_flush() the inode to try and force delalloc to happen, or as a last resort actually start allocation on all delalloc inodes in the fs. This has survived dbench, fs_mark and an fsx torture test. Signed-off-by: Josef Bacik <jbacik@redhat.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-09-12 00:12:44 +04:00
btrfs_unreserve_metadata_for_delalloc(root, inode, 1);
out_nolock:
kfree(pages);
if (pinned[0])
page_cache_release(pinned[0]);
if (pinned[1])
page_cache_release(pinned[1]);
*ppos = pos;
Btrfs: add extra flushing for renames and truncates Renames and truncates are both common ways to replace old data with new data. The filesystem can make an effort to make sure the new data is on disk before actually replacing the old data. This is especially important for rename, which many application use as though it were atomic for both the data and the metadata involved. The current btrfs code will happily replace a file that is fully on disk with one that was just created and still has pending IO. If we crash after transaction commit but before the IO is done, we'll end up replacing a good file with a zero length file. The solution used here is to create a list of inodes that need special ordering and force them to disk before the commit is done. This is similar to the ext3 style data=ordering, except it is only done on selected files. Btrfs is able to get away with this because it does not wait on commits very often, even for fsync (which use a sub-commit). For renames, we order the file when it wasn't already on disk and when it is replacing an existing file. Larger files are sent to filemap_flush right away (before the transaction handle is opened). For truncates, we order if the file goes from non-zero size down to zero size. This is a little different, because at the time of the truncate the file has no dirty bytes to order. But, we flag the inode so that it is added to the ordered list on close (via release method). We also immediately add it to the ordered list of the current transaction so that we can try to flush down any writes the application sneaks in before commit. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-03-31 21:27:11 +04:00
/*
* we want to make sure fsync finds this change
* but we haven't joined a transaction running right now.
*
* Later on, someone is sure to update the inode and get the
* real transid recorded.
*
* We set last_trans now to the fs_info generation + 1,
* this will either be one more than the running transaction
* or the generation used for the next transaction if there isn't
* one running right now.
*/
BTRFS_I(inode)->last_trans = root->fs_info->generation + 1;
if (num_written > 0 && will_write) {
struct btrfs_trans_handle *trans;
err = btrfs_wait_ordered_range(inode, start_pos, num_written);
if (err)
num_written = err;
if ((file->f_flags & O_SYNC) || IS_SYNC(inode)) {
trans = btrfs_start_transaction(root, 1);
ret = btrfs_log_dentry_safe(trans, root,
file->f_dentry);
if (ret == 0) {
Btrfs: tree logging unlink/rename fixes The tree logging code allows individual files or directories to be logged without including operations on other files and directories in the FS. It tries to commit the minimal set of changes to disk in order to fsync the single file or directory that was sent to fsync or O_SYNC. The tree logging code was allowing files and directories to be unlinked if they were part of a rename operation where only one directory in the rename was in the fsync log. This patch adds a few new rules to the tree logging. 1) on rename or unlink, if the inode being unlinked isn't in the fsync log, we must force a full commit before doing an fsync of the directory where the unlink was done. The commit isn't done during the unlink, but it is forced the next time we try to log the parent directory. Solution: record transid of last unlink/rename per directory when the directory wasn't already logged. For renames this is only done when renaming to a different directory. mkdir foo/some_dir normal commit rename foo/some_dir foo2/some_dir mkdir foo/some_dir fsync foo/some_dir/some_file The fsync above will unlink the original some_dir without recording it in its new location (foo2). After a crash, some_dir will be gone unless the fsync of some_file forces a full commit 2) we must log any new names for any file or dir that is in the fsync log. This way we make sure not to lose files that are unlinked during the same transaction. 2a) we must log any new names for any file or dir during rename when the directory they are being removed from was logged. 2a is actually the more important variant. Without the extra logging a crash might unlink the old name without recreating the new one 3) after a crash, we must go through any directories with a link count of zero and redo the rm -rf mkdir f1/foo normal commit rm -rf f1/foo fsync(f1) The directory f1 was fully removed from the FS, but fsync was never called on f1, only its parent dir. After a crash the rm -rf must be replayed. This must be able to recurse down the entire directory tree. The inode link count fixup code takes care of the ugly details. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-03-24 17:24:20 +03:00
ret = btrfs_sync_log(trans, root);
if (ret == 0)
btrfs_end_transaction(trans, root);
else
btrfs_commit_transaction(trans, root);
} else {
btrfs_commit_transaction(trans, root);
}
}
if (file->f_flags & O_DIRECT) {
invalidate_mapping_pages(inode->i_mapping,
start_pos >> PAGE_CACHE_SHIFT,
(start_pos + num_written - 1) >> PAGE_CACHE_SHIFT);
}
}
current->backing_dev_info = NULL;
return num_written ? num_written : err;
}
int btrfs_release_file(struct inode *inode, struct file *filp)
{
Btrfs: add extra flushing for renames and truncates Renames and truncates are both common ways to replace old data with new data. The filesystem can make an effort to make sure the new data is on disk before actually replacing the old data. This is especially important for rename, which many application use as though it were atomic for both the data and the metadata involved. The current btrfs code will happily replace a file that is fully on disk with one that was just created and still has pending IO. If we crash after transaction commit but before the IO is done, we'll end up replacing a good file with a zero length file. The solution used here is to create a list of inodes that need special ordering and force them to disk before the commit is done. This is similar to the ext3 style data=ordering, except it is only done on selected files. Btrfs is able to get away with this because it does not wait on commits very often, even for fsync (which use a sub-commit). For renames, we order the file when it wasn't already on disk and when it is replacing an existing file. Larger files are sent to filemap_flush right away (before the transaction handle is opened). For truncates, we order if the file goes from non-zero size down to zero size. This is a little different, because at the time of the truncate the file has no dirty bytes to order. But, we flag the inode so that it is added to the ordered list on close (via release method). We also immediately add it to the ordered list of the current transaction so that we can try to flush down any writes the application sneaks in before commit. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-03-31 21:27:11 +04:00
/*
* ordered_data_close is set by settattr when we are about to truncate
* a file from a non-zero size to a zero size. This tries to
* flush down new bytes that may have been written if the
* application were using truncate to replace a file in place.
*/
if (BTRFS_I(inode)->ordered_data_close) {
BTRFS_I(inode)->ordered_data_close = 0;
btrfs_add_ordered_operation(NULL, BTRFS_I(inode)->root, inode);
if (inode->i_size > BTRFS_ORDERED_OPERATIONS_FLUSH_LIMIT)
filemap_flush(inode->i_mapping);
}
if (filp->private_data)
btrfs_ioctl_trans_end(filp);
return 0;
}
/*
* fsync call for both files and directories. This logs the inode into
* the tree log instead of forcing full commits whenever possible.
*
* It needs to call filemap_fdatawait so that all ordered extent updates are
* in the metadata btree are up to date for copying to the log.
*
* It drops the inode mutex before doing the tree log commit. This is an
* important optimization for directories because holding the mutex prevents
* new operations on the dir while we write to disk.
*/
int btrfs_sync_file(struct file *file, struct dentry *dentry, int datasync)
{
struct inode *inode = dentry->d_inode;
struct btrfs_root *root = BTRFS_I(inode)->root;
int ret = 0;
struct btrfs_trans_handle *trans;
/*
* check the transaction that last modified this inode
* and see if its already been committed
*/
if (!BTRFS_I(inode)->last_trans)
goto out;
mutex_lock(&root->fs_info->trans_mutex);
if (BTRFS_I(inode)->last_trans <=
root->fs_info->last_trans_committed) {
BTRFS_I(inode)->last_trans = 0;
mutex_unlock(&root->fs_info->trans_mutex);
goto out;
}
mutex_unlock(&root->fs_info->trans_mutex);
root->log_batch++;
filemap_fdatawrite(inode->i_mapping);
btrfs_wait_ordered_range(inode, 0, (u64)-1);
root->log_batch++;
if (datasync && !(inode->i_state & I_DIRTY_PAGES))
goto out;
/*
* ok we haven't committed the transaction yet, lets do a commit
*/
if (file && file->private_data)
btrfs_ioctl_trans_end(file);
trans = btrfs_start_transaction(root, 1);
if (!trans) {
ret = -ENOMEM;
goto out;
}
ret = btrfs_log_dentry_safe(trans, root, dentry);
if (ret < 0)
goto out;
/* we've logged all the items and now have a consistent
* version of the file in the log. It is possible that
* someone will come in and modify the file, but that's
* fine because the log is consistent on disk, and we
* have references to all of the file's extents
*
* It is possible that someone will come in and log the
* file again, but that will end up using the synchronization
* inside btrfs_sync_log to keep things safe.
*/
mutex_unlock(&dentry->d_inode->i_mutex);
if (ret > 0) {
ret = btrfs_commit_transaction(trans, root);
} else {
Btrfs: tree logging unlink/rename fixes The tree logging code allows individual files or directories to be logged without including operations on other files and directories in the FS. It tries to commit the minimal set of changes to disk in order to fsync the single file or directory that was sent to fsync or O_SYNC. The tree logging code was allowing files and directories to be unlinked if they were part of a rename operation where only one directory in the rename was in the fsync log. This patch adds a few new rules to the tree logging. 1) on rename or unlink, if the inode being unlinked isn't in the fsync log, we must force a full commit before doing an fsync of the directory where the unlink was done. The commit isn't done during the unlink, but it is forced the next time we try to log the parent directory. Solution: record transid of last unlink/rename per directory when the directory wasn't already logged. For renames this is only done when renaming to a different directory. mkdir foo/some_dir normal commit rename foo/some_dir foo2/some_dir mkdir foo/some_dir fsync foo/some_dir/some_file The fsync above will unlink the original some_dir without recording it in its new location (foo2). After a crash, some_dir will be gone unless the fsync of some_file forces a full commit 2) we must log any new names for any file or dir that is in the fsync log. This way we make sure not to lose files that are unlinked during the same transaction. 2a) we must log any new names for any file or dir during rename when the directory they are being removed from was logged. 2a is actually the more important variant. Without the extra logging a crash might unlink the old name without recreating the new one 3) after a crash, we must go through any directories with a link count of zero and redo the rm -rf mkdir f1/foo normal commit rm -rf f1/foo fsync(f1) The directory f1 was fully removed from the FS, but fsync was never called on f1, only its parent dir. After a crash the rm -rf must be replayed. This must be able to recurse down the entire directory tree. The inode link count fixup code takes care of the ugly details. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-03-24 17:24:20 +03:00
ret = btrfs_sync_log(trans, root);
if (ret == 0)
ret = btrfs_end_transaction(trans, root);
else
ret = btrfs_commit_transaction(trans, root);
}
mutex_lock(&dentry->d_inode->i_mutex);
out:
return ret > 0 ? EIO : ret;
}
static struct vm_operations_struct btrfs_file_vm_ops = {
.fault = filemap_fault,
.page_mkwrite = btrfs_page_mkwrite,
};
static int btrfs_file_mmap(struct file *filp, struct vm_area_struct *vma)
{
vma->vm_ops = &btrfs_file_vm_ops;
file_accessed(filp);
return 0;
}
struct file_operations btrfs_file_operations = {
.llseek = generic_file_llseek,
.read = do_sync_read,
.aio_read = generic_file_aio_read,
.splice_read = generic_file_splice_read,
.write = btrfs_file_write,
.mmap = btrfs_file_mmap,
.open = generic_file_open,
.release = btrfs_release_file,
.fsync = btrfs_sync_file,
.unlocked_ioctl = btrfs_ioctl,
#ifdef CONFIG_COMPAT
.compat_ioctl = btrfs_ioctl,
#endif
};