linux_dsm_epyc7002/fs/btrfs/compression.c

1137 lines
29 KiB
C
Raw Normal View History

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>
2008-10-30 01:49:59 +07:00
/*
* Copyright (C) 2008 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/kernel.h>
#include <linux/bio.h>
#include <linux/buffer_head.h>
#include <linux/file.h>
#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/bit_spinlock.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 15:04:11 +07:00
#include <linux/slab.h>
Btrfs: Add zlib compression support This is a large change for adding compression on reading and writing, both for inline and regular extents. It does some fairly large surgery to the writeback paths. Compression is off by default and enabled by mount -o compress. Even when the -o compress mount option is not used, it is possible to read compressed extents off the disk. If compression for a given set of pages fails to make them smaller, the file is flagged to avoid future compression attempts later. * While finding delalloc extents, the pages are locked before being sent down to the delalloc handler. This allows the delalloc handler to do complex things such as cleaning the pages, marking them writeback and starting IO on their behalf. * Inline extents are inserted at delalloc time now. This allows us to compress the data before inserting the inline extent, and it allows us to insert an inline extent that spans multiple pages. * All of the in-memory extent representations (extent_map.c, ordered-data.c etc) are changed to record both an in-memory size and an on disk size, as well as a flag for compression. From a disk format point of view, the extent pointers in the file are changed to record the on disk size of a given extent and some encoding flags. Space in the disk format is allocated for compression encoding, as well as encryption and a generic 'other' field. Neither the encryption or the 'other' field are currently used. In order to limit the amount of data read for a single random read in the file, the size of a compressed extent is limited to 128k. This is a software only limit, the disk format supports u64 sized compressed extents. In order to limit the ram consumed while processing extents, the uncompressed size of a compressed extent is limited to 256k. This is a software only limit and will be subject to tuning later. Checksumming is still done on compressed extents, and it is done on the uncompressed version of the data. This way additional encodings can be layered on without having to figure out which encoding to checksum. Compression happens at delalloc time, which is basically singled threaded because it is usually done by a single pdflush thread. This makes it tricky to spread the compression load across all the cpus on the box. We'll have to look at parallel pdflush walks of dirty inodes at a later time. Decompression is hooked into readpages and it does spread across CPUs nicely. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-10-30 01:49:59 +07:00
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
#include "volumes.h"
#include "ordered-data.h"
#include "compression.h"
#include "extent_io.h"
#include "extent_map.h"
struct compressed_bio {
/* number of bios pending for this compressed extent */
atomic_t pending_bios;
/* the pages with the compressed data on them */
struct page **compressed_pages;
/* inode that owns this data */
struct inode *inode;
/* starting offset in the inode for our pages */
u64 start;
/* number of bytes in the inode we're working on */
unsigned long len;
/* number of bytes on disk */
unsigned long compressed_len;
/* the compression algorithm for this bio */
int compress_type;
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>
2008-10-30 01:49:59 +07:00
/* number of compressed pages in the array */
unsigned long nr_pages;
/* IO errors */
int errors;
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
int mirror_num;
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>
2008-10-30 01:49:59 +07:00
/* for reads, this is the bio we are copying the data into */
struct bio *orig_bio;
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
/*
* the start of a variable length array of checksums only
* used by reads
*/
u32 sums;
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>
2008-10-30 01:49:59 +07:00
};
static int btrfs_decompress_biovec(int type, struct page **pages_in,
u64 disk_start, struct bio_vec *bvec,
int vcnt, size_t srclen);
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
static inline int compressed_bio_size(struct btrfs_root *root,
unsigned long disk_size)
{
u16 csum_size = btrfs_super_csum_size(root->fs_info->super_copy);
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
return sizeof(struct compressed_bio) +
(DIV_ROUND_UP(disk_size, root->sectorsize)) * csum_size;
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
}
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>
2008-10-30 01:49:59 +07:00
static struct bio *compressed_bio_alloc(struct block_device *bdev,
u64 first_byte, gfp_t gfp_flags)
{
return btrfs_bio_alloc(bdev, first_byte >> 9, BIO_MAX_PAGES, gfp_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>
2008-10-30 01:49:59 +07:00
}
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
static int check_compressed_csum(struct inode *inode,
struct compressed_bio *cb,
u64 disk_start)
{
int ret;
struct page *page;
unsigned long i;
char *kaddr;
u32 csum;
u32 *cb_sum = &cb->sums;
if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
return 0;
for (i = 0; i < cb->nr_pages; i++) {
page = cb->compressed_pages[i];
csum = ~(u32)0;
kaddr = kmap_atomic(page);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
csum = btrfs_csum_data(kaddr, csum, PAGE_SIZE);
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
btrfs_csum_final(csum, (char *)&csum);
kunmap_atomic(kaddr);
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
if (csum != *cb_sum) {
btrfs_info(BTRFS_I(inode)->root->fs_info,
"csum failed ino %llu extent %llu csum %u wanted %u mirror %d",
btrfs_ino(inode), disk_start, csum, *cb_sum,
cb->mirror_num);
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
ret = -EIO;
goto fail;
}
cb_sum++;
}
ret = 0;
fail:
return 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>
2008-10-30 01:49:59 +07:00
/* when we finish reading compressed pages from the disk, we
* decompress them and then run the bio end_io routines on the
* decompressed pages (in the inode address space).
*
* This allows the checksumming and other IO error handling routines
* to work normally
*
* The compressed pages are freed here, and it must be run
* in process context
*/
static void end_compressed_bio_read(struct bio *bio)
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>
2008-10-30 01:49:59 +07:00
{
struct compressed_bio *cb = bio->bi_private;
struct inode *inode;
struct page *page;
unsigned long index;
int ret;
if (bio->bi_error)
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>
2008-10-30 01:49:59 +07:00
cb->errors = 1;
/* if there are more bios still pending for this compressed
* extent, just exit
*/
if (!atomic_dec_and_test(&cb->pending_bios))
goto out;
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
inode = cb->inode;
block: Abstract out bvec iterator Immutable biovecs are going to require an explicit iterator. To implement immutable bvecs, a later patch is going to add a bi_bvec_done member to this struct; for now, this patch effectively just renames things. Signed-off-by: Kent Overstreet <kmo@daterainc.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: "Ed L. Cashin" <ecashin@coraid.com> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Lars Ellenberg <drbd-dev@lists.linbit.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Matthew Wilcox <willy@linux.intel.com> Cc: Geoff Levand <geoff@infradead.org> Cc: Yehuda Sadeh <yehuda@inktank.com> Cc: Sage Weil <sage@inktank.com> Cc: Alex Elder <elder@inktank.com> Cc: ceph-devel@vger.kernel.org Cc: Joshua Morris <josh.h.morris@us.ibm.com> Cc: Philip Kelleher <pjk1939@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Neil Brown <neilb@suse.de> Cc: Alasdair Kergon <agk@redhat.com> Cc: Mike Snitzer <snitzer@redhat.com> Cc: dm-devel@redhat.com Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: linux390@de.ibm.com Cc: Boaz Harrosh <bharrosh@panasas.com> Cc: Benny Halevy <bhalevy@tonian.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Chris Mason <chris.mason@fusionio.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: Jaegeuk Kim <jaegeuk.kim@samsung.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@kernel.org> Cc: Joern Engel <joern@logfs.org> Cc: Prasad Joshi <prasadjoshi.linux@gmail.com> Cc: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: KONISHI Ryusuke <konishi.ryusuke@lab.ntt.co.jp> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Ben Myers <bpm@sgi.com> Cc: xfs@oss.sgi.com Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Len Brown <len.brown@intel.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Herton Ronaldo Krzesinski <herton.krzesinski@canonical.com> Cc: Ben Hutchings <ben@decadent.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Guo Chao <yan@linux.vnet.ibm.com> Cc: Tejun Heo <tj@kernel.org> Cc: Asai Thambi S P <asamymuthupa@micron.com> Cc: Selvan Mani <smani@micron.com> Cc: Sam Bradshaw <sbradshaw@micron.com> Cc: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Cc: "Roger Pau Monné" <roger.pau@citrix.com> Cc: Jan Beulich <jbeulich@suse.com> Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Cc: Ian Campbell <Ian.Campbell@citrix.com> Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Jiang Liu <jiang.liu@huawei.com> Cc: Nitin Gupta <ngupta@vflare.org> Cc: Jerome Marchand <jmarchand@redhat.com> Cc: Joe Perches <joe@perches.com> Cc: Peng Tao <tao.peng@emc.com> Cc: Andy Adamson <andros@netapp.com> Cc: fanchaoting <fanchaoting@cn.fujitsu.com> Cc: Jie Liu <jeff.liu@oracle.com> Cc: Sunil Mushran <sunil.mushran@gmail.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: Namjae Jeon <namjae.jeon@samsung.com> Cc: Pankaj Kumar <pankaj.km@samsung.com> Cc: Dan Magenheimer <dan.magenheimer@oracle.com> Cc: Mel Gorman <mgorman@suse.de>6
2013-10-12 05:44:27 +07:00
ret = check_compressed_csum(inode, cb,
(u64)bio->bi_iter.bi_sector << 9);
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
if (ret)
goto csum_failed;
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>
2008-10-30 01:49:59 +07:00
/* ok, we're the last bio for this extent, lets start
* the decompression.
*/
ret = btrfs_decompress_biovec(cb->compress_type,
cb->compressed_pages,
cb->start,
cb->orig_bio->bi_io_vec,
cb->orig_bio->bi_vcnt,
cb->compressed_len);
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
csum_failed:
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>
2008-10-30 01:49:59 +07:00
if (ret)
cb->errors = 1;
/* release the compressed pages */
index = 0;
for (index = 0; index < cb->nr_pages; index++) {
page = cb->compressed_pages[index];
page->mapping = NULL;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
put_page(page);
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>
2008-10-30 01:49:59 +07:00
}
/* do io completion on the original bio */
if (cb->errors) {
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>
2008-10-30 01:49:59 +07:00
bio_io_error(cb->orig_bio);
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
} else {
int i;
struct bio_vec *bvec;
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
/*
* we have verified the checksum already, set page
* checked so the end_io handlers know about it
*/
bio_for_each_segment_all(bvec, cb->orig_bio, i)
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
SetPageChecked(bvec->bv_page);
bio_endio(cb->orig_bio);
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
}
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>
2008-10-30 01:49:59 +07:00
/* finally free the cb struct */
kfree(cb->compressed_pages);
kfree(cb);
out:
bio_put(bio);
}
/*
* Clear the writeback bits on all of the file
* pages for a compressed write
*/
static noinline void end_compressed_writeback(struct inode *inode,
const struct compressed_bio *cb)
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>
2008-10-30 01:49:59 +07:00
{
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
unsigned long index = cb->start >> PAGE_SHIFT;
unsigned long end_index = (cb->start + cb->len - 1) >> PAGE_SHIFT;
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>
2008-10-30 01:49:59 +07:00
struct page *pages[16];
unsigned long nr_pages = end_index - index + 1;
int i;
int ret;
if (cb->errors)
mapping_set_error(inode->i_mapping, -EIO);
while (nr_pages > 0) {
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>
2008-10-30 01:49:59 +07:00
ret = find_get_pages_contig(inode->i_mapping, index,
min_t(unsigned long,
nr_pages, ARRAY_SIZE(pages)), pages);
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>
2008-10-30 01:49:59 +07:00
if (ret == 0) {
nr_pages -= 1;
index += 1;
continue;
}
for (i = 0; i < ret; i++) {
if (cb->errors)
SetPageError(pages[i]);
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>
2008-10-30 01:49:59 +07:00
end_page_writeback(pages[i]);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
put_page(pages[i]);
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>
2008-10-30 01:49:59 +07:00
}
nr_pages -= ret;
index += ret;
}
/* the inode may be gone now */
}
/*
* do the cleanup once all the compressed pages hit the disk.
* This will clear writeback on the file pages and free the compressed
* pages.
*
* This also calls the writeback end hooks for the file pages so that
* metadata and checksums can be updated in the file.
*/
static void end_compressed_bio_write(struct bio *bio)
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>
2008-10-30 01:49:59 +07:00
{
struct extent_io_tree *tree;
struct compressed_bio *cb = bio->bi_private;
struct inode *inode;
struct page *page;
unsigned long index;
if (bio->bi_error)
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>
2008-10-30 01:49:59 +07:00
cb->errors = 1;
/* if there are more bios still pending for this compressed
* extent, just exit
*/
if (!atomic_dec_and_test(&cb->pending_bios))
goto out;
/* ok, we're the last bio for this extent, step one is to
* call back into the FS and do all the end_io operations
*/
inode = cb->inode;
tree = &BTRFS_I(inode)->io_tree;
cb->compressed_pages[0]->mapping = cb->inode->i_mapping;
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>
2008-10-30 01:49:59 +07:00
tree->ops->writepage_end_io_hook(cb->compressed_pages[0],
cb->start,
cb->start + cb->len - 1,
NULL,
bio->bi_error ? 0 : 1);
cb->compressed_pages[0]->mapping = NULL;
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>
2008-10-30 01:49:59 +07:00
end_compressed_writeback(inode, cb);
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>
2008-10-30 01:49:59 +07:00
/* note, our inode could be gone now */
/*
* release the compressed pages, these came from alloc_page and
* are not attached to the inode at all
*/
index = 0;
for (index = 0; index < cb->nr_pages; index++) {
page = cb->compressed_pages[index];
page->mapping = NULL;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
put_page(page);
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>
2008-10-30 01:49:59 +07:00
}
/* finally free the cb struct */
kfree(cb->compressed_pages);
kfree(cb);
out:
bio_put(bio);
}
/*
* worker function to build and submit bios for previously compressed pages.
* The corresponding pages in the inode should be marked for writeback
* and the compressed pages should have a reference on them for dropping
* when the IO is complete.
*
* This also checksums the file bytes and gets things ready for
* the end io hooks.
*/
int btrfs_submit_compressed_write(struct inode *inode, u64 start,
unsigned long len, u64 disk_start,
unsigned long compressed_len,
struct page **compressed_pages,
unsigned long nr_pages)
{
struct bio *bio = NULL;
struct btrfs_root *root = BTRFS_I(inode)->root;
struct compressed_bio *cb;
unsigned long bytes_left;
struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
int pg_index = 0;
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>
2008-10-30 01:49:59 +07:00
struct page *page;
u64 first_byte = disk_start;
struct block_device *bdev;
int ret;
int skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
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>
2008-10-30 01:49:59 +07:00
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
WARN_ON(start & ((u64)PAGE_SIZE - 1));
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
cb = kmalloc(compressed_bio_size(root, compressed_len), GFP_NOFS);
if (!cb)
return -ENOMEM;
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>
2008-10-30 01:49:59 +07:00
atomic_set(&cb->pending_bios, 0);
cb->errors = 0;
cb->inode = inode;
cb->start = start;
cb->len = len;
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
cb->mirror_num = 0;
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>
2008-10-30 01:49:59 +07:00
cb->compressed_pages = compressed_pages;
cb->compressed_len = compressed_len;
cb->orig_bio = NULL;
cb->nr_pages = nr_pages;
bdev = BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev;
bio = compressed_bio_alloc(bdev, first_byte, GFP_NOFS);
if (!bio) {
kfree(cb);
return -ENOMEM;
}
bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
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>
2008-10-30 01:49:59 +07:00
bio->bi_private = cb;
bio->bi_end_io = end_compressed_bio_write;
atomic_inc(&cb->pending_bios);
/* create and submit bios for the compressed pages */
bytes_left = compressed_len;
for (pg_index = 0; pg_index < cb->nr_pages; pg_index++) {
page = compressed_pages[pg_index];
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>
2008-10-30 01:49:59 +07:00
page->mapping = inode->i_mapping;
block: Abstract out bvec iterator Immutable biovecs are going to require an explicit iterator. To implement immutable bvecs, a later patch is going to add a bi_bvec_done member to this struct; for now, this patch effectively just renames things. Signed-off-by: Kent Overstreet <kmo@daterainc.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: "Ed L. Cashin" <ecashin@coraid.com> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Lars Ellenberg <drbd-dev@lists.linbit.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Matthew Wilcox <willy@linux.intel.com> Cc: Geoff Levand <geoff@infradead.org> Cc: Yehuda Sadeh <yehuda@inktank.com> Cc: Sage Weil <sage@inktank.com> Cc: Alex Elder <elder@inktank.com> Cc: ceph-devel@vger.kernel.org Cc: Joshua Morris <josh.h.morris@us.ibm.com> Cc: Philip Kelleher <pjk1939@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Neil Brown <neilb@suse.de> Cc: Alasdair Kergon <agk@redhat.com> Cc: Mike Snitzer <snitzer@redhat.com> Cc: dm-devel@redhat.com Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: linux390@de.ibm.com Cc: Boaz Harrosh <bharrosh@panasas.com> Cc: Benny Halevy <bhalevy@tonian.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Chris Mason <chris.mason@fusionio.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: Jaegeuk Kim <jaegeuk.kim@samsung.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@kernel.org> Cc: Joern Engel <joern@logfs.org> Cc: Prasad Joshi <prasadjoshi.linux@gmail.com> Cc: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: KONISHI Ryusuke <konishi.ryusuke@lab.ntt.co.jp> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Ben Myers <bpm@sgi.com> Cc: xfs@oss.sgi.com Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Len Brown <len.brown@intel.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Herton Ronaldo Krzesinski <herton.krzesinski@canonical.com> Cc: Ben Hutchings <ben@decadent.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Guo Chao <yan@linux.vnet.ibm.com> Cc: Tejun Heo <tj@kernel.org> Cc: Asai Thambi S P <asamymuthupa@micron.com> Cc: Selvan Mani <smani@micron.com> Cc: Sam Bradshaw <sbradshaw@micron.com> Cc: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Cc: "Roger Pau Monné" <roger.pau@citrix.com> Cc: Jan Beulich <jbeulich@suse.com> Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Cc: Ian Campbell <Ian.Campbell@citrix.com> Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Jiang Liu <jiang.liu@huawei.com> Cc: Nitin Gupta <ngupta@vflare.org> Cc: Jerome Marchand <jmarchand@redhat.com> Cc: Joe Perches <joe@perches.com> Cc: Peng Tao <tao.peng@emc.com> Cc: Andy Adamson <andros@netapp.com> Cc: fanchaoting <fanchaoting@cn.fujitsu.com> Cc: Jie Liu <jeff.liu@oracle.com> Cc: Sunil Mushran <sunil.mushran@gmail.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: Namjae Jeon <namjae.jeon@samsung.com> Cc: Pankaj Kumar <pankaj.km@samsung.com> Cc: Dan Magenheimer <dan.magenheimer@oracle.com> Cc: Mel Gorman <mgorman@suse.de>6
2013-10-12 05:44:27 +07:00
if (bio->bi_iter.bi_size)
ret = io_tree->ops->merge_bio_hook(page, 0,
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
PAGE_SIZE,
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>
2008-10-30 01:49:59 +07:00
bio, 0);
else
ret = 0;
page->mapping = NULL;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
if (ret || bio_add_page(bio, page, PAGE_SIZE, 0) <
PAGE_SIZE) {
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>
2008-10-30 01:49:59 +07:00
bio_get(bio);
/*
* inc the count before we submit the bio so
* we know the end IO handler won't happen before
* we inc the count. Otherwise, the cb might get
* freed before we're done setting it up
*/
atomic_inc(&cb->pending_bios);
ret = btrfs_bio_wq_end_io(root->fs_info, bio,
BTRFS_WQ_ENDIO_DATA);
BUG_ON(ret); /* -ENOMEM */
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>
2008-10-30 01:49:59 +07:00
if (!skip_sum) {
ret = btrfs_csum_one_bio(root, inode, bio,
start, 1);
BUG_ON(ret); /* -ENOMEM */
}
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
ret = btrfs_map_bio(root, bio, 0, 1);
if (ret) {
bio->bi_error = ret;
bio_endio(bio);
}
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>
2008-10-30 01:49:59 +07:00
bio_put(bio);
bio = compressed_bio_alloc(bdev, first_byte, GFP_NOFS);
BUG_ON(!bio);
bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
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>
2008-10-30 01:49:59 +07:00
bio->bi_private = cb;
bio->bi_end_io = end_compressed_bio_write;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
bio_add_page(bio, page, PAGE_SIZE, 0);
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>
2008-10-30 01:49:59 +07:00
}
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
if (bytes_left < PAGE_SIZE) {
btrfs_info(BTRFS_I(inode)->root->fs_info,
"bytes left %lu compress len %lu nr %lu",
bytes_left, cb->compressed_len, cb->nr_pages);
}
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
bytes_left -= PAGE_SIZE;
first_byte += PAGE_SIZE;
cond_resched();
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>
2008-10-30 01:49:59 +07:00
}
bio_get(bio);
ret = btrfs_bio_wq_end_io(root->fs_info, bio, BTRFS_WQ_ENDIO_DATA);
BUG_ON(ret); /* -ENOMEM */
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>
2008-10-30 01:49:59 +07:00
if (!skip_sum) {
ret = btrfs_csum_one_bio(root, inode, bio, start, 1);
BUG_ON(ret); /* -ENOMEM */
}
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
ret = btrfs_map_bio(root, bio, 0, 1);
if (ret) {
bio->bi_error = ret;
bio_endio(bio);
}
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>
2008-10-30 01:49:59 +07:00
bio_put(bio);
return 0;
}
static noinline int add_ra_bio_pages(struct inode *inode,
u64 compressed_end,
struct compressed_bio *cb)
{
unsigned long end_index;
unsigned long pg_index;
u64 last_offset;
u64 isize = i_size_read(inode);
int ret;
struct page *page;
unsigned long nr_pages = 0;
struct extent_map *em;
struct address_space *mapping = inode->i_mapping;
struct extent_map_tree *em_tree;
struct extent_io_tree *tree;
u64 end;
int misses = 0;
page = cb->orig_bio->bi_io_vec[cb->orig_bio->bi_vcnt - 1].bv_page;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
last_offset = (page_offset(page) + PAGE_SIZE);
em_tree = &BTRFS_I(inode)->extent_tree;
tree = &BTRFS_I(inode)->io_tree;
if (isize == 0)
return 0;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
end_index = (i_size_read(inode) - 1) >> PAGE_SHIFT;
while (last_offset < compressed_end) {
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
pg_index = last_offset >> PAGE_SHIFT;
if (pg_index > end_index)
break;
rcu_read_lock();
page = radix_tree_lookup(&mapping->page_tree, pg_index);
rcu_read_unlock();
2014-04-04 04:47:46 +07:00
if (page && !radix_tree_exceptional_entry(page)) {
misses++;
if (misses > 4)
break;
goto next;
}
page = __page_cache_alloc(mapping_gfp_constraint(mapping,
~__GFP_FS));
if (!page)
break;
if (add_to_page_cache_lru(page, mapping, pg_index, GFP_NOFS)) {
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
put_page(page);
goto next;
}
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
end = last_offset + PAGE_SIZE - 1;
/*
* at this point, we have a locked page in the page cache
* for these bytes in the file. But, we have to make
* sure they map to this compressed extent on disk.
*/
set_page_extent_mapped(page);
lock_extent(tree, last_offset, end);
read_lock(&em_tree->lock);
em = lookup_extent_mapping(em_tree, last_offset,
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
PAGE_SIZE);
read_unlock(&em_tree->lock);
if (!em || last_offset < em->start ||
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
(last_offset + PAGE_SIZE > extent_map_end(em)) ||
block: Abstract out bvec iterator Immutable biovecs are going to require an explicit iterator. To implement immutable bvecs, a later patch is going to add a bi_bvec_done member to this struct; for now, this patch effectively just renames things. Signed-off-by: Kent Overstreet <kmo@daterainc.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: "Ed L. Cashin" <ecashin@coraid.com> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Lars Ellenberg <drbd-dev@lists.linbit.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Matthew Wilcox <willy@linux.intel.com> Cc: Geoff Levand <geoff@infradead.org> Cc: Yehuda Sadeh <yehuda@inktank.com> Cc: Sage Weil <sage@inktank.com> Cc: Alex Elder <elder@inktank.com> Cc: ceph-devel@vger.kernel.org Cc: Joshua Morris <josh.h.morris@us.ibm.com> Cc: Philip Kelleher <pjk1939@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Neil Brown <neilb@suse.de> Cc: Alasdair Kergon <agk@redhat.com> Cc: Mike Snitzer <snitzer@redhat.com> Cc: dm-devel@redhat.com Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: linux390@de.ibm.com Cc: Boaz Harrosh <bharrosh@panasas.com> Cc: Benny Halevy <bhalevy@tonian.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Chris Mason <chris.mason@fusionio.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: Jaegeuk Kim <jaegeuk.kim@samsung.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@kernel.org> Cc: Joern Engel <joern@logfs.org> Cc: Prasad Joshi <prasadjoshi.linux@gmail.com> Cc: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: KONISHI Ryusuke <konishi.ryusuke@lab.ntt.co.jp> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Ben Myers <bpm@sgi.com> Cc: xfs@oss.sgi.com Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Len Brown <len.brown@intel.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Herton Ronaldo Krzesinski <herton.krzesinski@canonical.com> Cc: Ben Hutchings <ben@decadent.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Guo Chao <yan@linux.vnet.ibm.com> Cc: Tejun Heo <tj@kernel.org> Cc: Asai Thambi S P <asamymuthupa@micron.com> Cc: Selvan Mani <smani@micron.com> Cc: Sam Bradshaw <sbradshaw@micron.com> Cc: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Cc: "Roger Pau Monné" <roger.pau@citrix.com> Cc: Jan Beulich <jbeulich@suse.com> Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Cc: Ian Campbell <Ian.Campbell@citrix.com> Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Jiang Liu <jiang.liu@huawei.com> Cc: Nitin Gupta <ngupta@vflare.org> Cc: Jerome Marchand <jmarchand@redhat.com> Cc: Joe Perches <joe@perches.com> Cc: Peng Tao <tao.peng@emc.com> Cc: Andy Adamson <andros@netapp.com> Cc: fanchaoting <fanchaoting@cn.fujitsu.com> Cc: Jie Liu <jeff.liu@oracle.com> Cc: Sunil Mushran <sunil.mushran@gmail.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: Namjae Jeon <namjae.jeon@samsung.com> Cc: Pankaj Kumar <pankaj.km@samsung.com> Cc: Dan Magenheimer <dan.magenheimer@oracle.com> Cc: Mel Gorman <mgorman@suse.de>6
2013-10-12 05:44:27 +07:00
(em->block_start >> 9) != cb->orig_bio->bi_iter.bi_sector) {
free_extent_map(em);
unlock_extent(tree, last_offset, end);
unlock_page(page);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
put_page(page);
break;
}
free_extent_map(em);
if (page->index == end_index) {
char *userpage;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
size_t zero_offset = isize & (PAGE_SIZE - 1);
if (zero_offset) {
int zeros;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
zeros = PAGE_SIZE - zero_offset;
userpage = kmap_atomic(page);
memset(userpage + zero_offset, 0, zeros);
flush_dcache_page(page);
kunmap_atomic(userpage);
}
}
ret = bio_add_page(cb->orig_bio, page,
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
PAGE_SIZE, 0);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
if (ret == PAGE_SIZE) {
nr_pages++;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
put_page(page);
} else {
unlock_extent(tree, last_offset, end);
unlock_page(page);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
put_page(page);
break;
}
next:
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
last_offset += PAGE_SIZE;
}
return 0;
}
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>
2008-10-30 01:49:59 +07:00
/*
* for a compressed read, the bio we get passed has all the inode pages
* in it. We don't actually do IO on those pages but allocate new ones
* to hold the compressed pages on disk.
*
block: Abstract out bvec iterator Immutable biovecs are going to require an explicit iterator. To implement immutable bvecs, a later patch is going to add a bi_bvec_done member to this struct; for now, this patch effectively just renames things. Signed-off-by: Kent Overstreet <kmo@daterainc.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: "Ed L. Cashin" <ecashin@coraid.com> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Lars Ellenberg <drbd-dev@lists.linbit.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Matthew Wilcox <willy@linux.intel.com> Cc: Geoff Levand <geoff@infradead.org> Cc: Yehuda Sadeh <yehuda@inktank.com> Cc: Sage Weil <sage@inktank.com> Cc: Alex Elder <elder@inktank.com> Cc: ceph-devel@vger.kernel.org Cc: Joshua Morris <josh.h.morris@us.ibm.com> Cc: Philip Kelleher <pjk1939@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Neil Brown <neilb@suse.de> Cc: Alasdair Kergon <agk@redhat.com> Cc: Mike Snitzer <snitzer@redhat.com> Cc: dm-devel@redhat.com Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: linux390@de.ibm.com Cc: Boaz Harrosh <bharrosh@panasas.com> Cc: Benny Halevy <bhalevy@tonian.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Chris Mason <chris.mason@fusionio.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: Jaegeuk Kim <jaegeuk.kim@samsung.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@kernel.org> Cc: Joern Engel <joern@logfs.org> Cc: Prasad Joshi <prasadjoshi.linux@gmail.com> Cc: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: KONISHI Ryusuke <konishi.ryusuke@lab.ntt.co.jp> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Ben Myers <bpm@sgi.com> Cc: xfs@oss.sgi.com Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Len Brown <len.brown@intel.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Herton Ronaldo Krzesinski <herton.krzesinski@canonical.com> Cc: Ben Hutchings <ben@decadent.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Guo Chao <yan@linux.vnet.ibm.com> Cc: Tejun Heo <tj@kernel.org> Cc: Asai Thambi S P <asamymuthupa@micron.com> Cc: Selvan Mani <smani@micron.com> Cc: Sam Bradshaw <sbradshaw@micron.com> Cc: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Cc: "Roger Pau Monné" <roger.pau@citrix.com> Cc: Jan Beulich <jbeulich@suse.com> Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Cc: Ian Campbell <Ian.Campbell@citrix.com> Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Jiang Liu <jiang.liu@huawei.com> Cc: Nitin Gupta <ngupta@vflare.org> Cc: Jerome Marchand <jmarchand@redhat.com> Cc: Joe Perches <joe@perches.com> Cc: Peng Tao <tao.peng@emc.com> Cc: Andy Adamson <andros@netapp.com> Cc: fanchaoting <fanchaoting@cn.fujitsu.com> Cc: Jie Liu <jeff.liu@oracle.com> Cc: Sunil Mushran <sunil.mushran@gmail.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: Namjae Jeon <namjae.jeon@samsung.com> Cc: Pankaj Kumar <pankaj.km@samsung.com> Cc: Dan Magenheimer <dan.magenheimer@oracle.com> Cc: Mel Gorman <mgorman@suse.de>6
2013-10-12 05:44:27 +07:00
* bio->bi_iter.bi_sector points to the compressed extent on disk
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>
2008-10-30 01:49:59 +07:00
* bio->bi_io_vec points to all of the inode pages
* bio->bi_vcnt is a count of pages
*
* After the compressed pages are read, we copy the bytes into the
* bio we were passed and then call the bio end_io calls
*/
int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
int mirror_num, unsigned long bio_flags)
{
struct extent_io_tree *tree;
struct extent_map_tree *em_tree;
struct compressed_bio *cb;
struct btrfs_root *root = BTRFS_I(inode)->root;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
unsigned long uncompressed_len = bio->bi_vcnt * PAGE_SIZE;
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>
2008-10-30 01:49:59 +07:00
unsigned long compressed_len;
unsigned long nr_pages;
unsigned long pg_index;
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>
2008-10-30 01:49:59 +07:00
struct page *page;
struct block_device *bdev;
struct bio *comp_bio;
block: Abstract out bvec iterator Immutable biovecs are going to require an explicit iterator. To implement immutable bvecs, a later patch is going to add a bi_bvec_done member to this struct; for now, this patch effectively just renames things. Signed-off-by: Kent Overstreet <kmo@daterainc.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: "Ed L. Cashin" <ecashin@coraid.com> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Lars Ellenberg <drbd-dev@lists.linbit.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Matthew Wilcox <willy@linux.intel.com> Cc: Geoff Levand <geoff@infradead.org> Cc: Yehuda Sadeh <yehuda@inktank.com> Cc: Sage Weil <sage@inktank.com> Cc: Alex Elder <elder@inktank.com> Cc: ceph-devel@vger.kernel.org Cc: Joshua Morris <josh.h.morris@us.ibm.com> Cc: Philip Kelleher <pjk1939@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Neil Brown <neilb@suse.de> Cc: Alasdair Kergon <agk@redhat.com> Cc: Mike Snitzer <snitzer@redhat.com> Cc: dm-devel@redhat.com Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: linux390@de.ibm.com Cc: Boaz Harrosh <bharrosh@panasas.com> Cc: Benny Halevy <bhalevy@tonian.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Chris Mason <chris.mason@fusionio.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: Jaegeuk Kim <jaegeuk.kim@samsung.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@kernel.org> Cc: Joern Engel <joern@logfs.org> Cc: Prasad Joshi <prasadjoshi.linux@gmail.com> Cc: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: KONISHI Ryusuke <konishi.ryusuke@lab.ntt.co.jp> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Ben Myers <bpm@sgi.com> Cc: xfs@oss.sgi.com Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Len Brown <len.brown@intel.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Herton Ronaldo Krzesinski <herton.krzesinski@canonical.com> Cc: Ben Hutchings <ben@decadent.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Guo Chao <yan@linux.vnet.ibm.com> Cc: Tejun Heo <tj@kernel.org> Cc: Asai Thambi S P <asamymuthupa@micron.com> Cc: Selvan Mani <smani@micron.com> Cc: Sam Bradshaw <sbradshaw@micron.com> Cc: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Cc: "Roger Pau Monné" <roger.pau@citrix.com> Cc: Jan Beulich <jbeulich@suse.com> Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Cc: Ian Campbell <Ian.Campbell@citrix.com> Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Jiang Liu <jiang.liu@huawei.com> Cc: Nitin Gupta <ngupta@vflare.org> Cc: Jerome Marchand <jmarchand@redhat.com> Cc: Joe Perches <joe@perches.com> Cc: Peng Tao <tao.peng@emc.com> Cc: Andy Adamson <andros@netapp.com> Cc: fanchaoting <fanchaoting@cn.fujitsu.com> Cc: Jie Liu <jeff.liu@oracle.com> Cc: Sunil Mushran <sunil.mushran@gmail.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: Namjae Jeon <namjae.jeon@samsung.com> Cc: Pankaj Kumar <pankaj.km@samsung.com> Cc: Dan Magenheimer <dan.magenheimer@oracle.com> Cc: Mel Gorman <mgorman@suse.de>6
2013-10-12 05:44:27 +07:00
u64 cur_disk_byte = (u64)bio->bi_iter.bi_sector << 9;
u64 em_len;
u64 em_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>
2008-10-30 01:49:59 +07:00
struct extent_map *em;
int ret = -ENOMEM;
int faili = 0;
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
u32 *sums;
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>
2008-10-30 01:49:59 +07:00
tree = &BTRFS_I(inode)->io_tree;
em_tree = &BTRFS_I(inode)->extent_tree;
/* we need the actual starting offset of this extent in the file */
read_lock(&em_tree->lock);
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>
2008-10-30 01:49:59 +07:00
em = lookup_extent_mapping(em_tree,
page_offset(bio->bi_io_vec->bv_page),
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
PAGE_SIZE);
read_unlock(&em_tree->lock);
if (!em)
return -EIO;
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>
2008-10-30 01:49:59 +07:00
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
compressed_len = em->block_len;
cb = kmalloc(compressed_bio_size(root, compressed_len), GFP_NOFS);
if (!cb)
goto out;
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>
2008-10-30 01:49:59 +07:00
atomic_set(&cb->pending_bios, 0);
cb->errors = 0;
cb->inode = inode;
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
cb->mirror_num = mirror_num;
sums = &cb->sums;
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>
2008-10-30 01:49:59 +07:00
cb->start = em->orig_start;
em_len = em->len;
em_start = em->start;
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
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>
2008-10-30 01:49:59 +07:00
free_extent_map(em);
em = NULL;
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>
2008-10-30 01:49:59 +07:00
cb->len = uncompressed_len;
cb->compressed_len = compressed_len;
cb->compress_type = extent_compress_type(bio_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>
2008-10-30 01:49:59 +07:00
cb->orig_bio = bio;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
nr_pages = DIV_ROUND_UP(compressed_len, PAGE_SIZE);
cb->compressed_pages = kcalloc(nr_pages, sizeof(struct page *),
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>
2008-10-30 01:49:59 +07:00
GFP_NOFS);
if (!cb->compressed_pages)
goto fail1;
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>
2008-10-30 01:49:59 +07:00
bdev = BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev;
for (pg_index = 0; pg_index < nr_pages; pg_index++) {
cb->compressed_pages[pg_index] = alloc_page(GFP_NOFS |
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>
2008-10-30 01:49:59 +07:00
__GFP_HIGHMEM);
if (!cb->compressed_pages[pg_index]) {
faili = pg_index - 1;
ret = -ENOMEM;
goto fail2;
}
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>
2008-10-30 01:49:59 +07:00
}
faili = nr_pages - 1;
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>
2008-10-30 01:49:59 +07:00
cb->nr_pages = nr_pages;
add_ra_bio_pages(inode, em_start + em_len, cb);
/* include any pages we added in add_ra-bio_pages */
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
uncompressed_len = bio->bi_vcnt * PAGE_SIZE;
cb->len = uncompressed_len;
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>
2008-10-30 01:49:59 +07:00
comp_bio = compressed_bio_alloc(bdev, cur_disk_byte, GFP_NOFS);
if (!comp_bio)
goto fail2;
bio_set_op_attrs (comp_bio, REQ_OP_READ, 0);
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>
2008-10-30 01:49:59 +07:00
comp_bio->bi_private = cb;
comp_bio->bi_end_io = end_compressed_bio_read;
atomic_inc(&cb->pending_bios);
for (pg_index = 0; pg_index < nr_pages; pg_index++) {
page = cb->compressed_pages[pg_index];
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>
2008-10-30 01:49:59 +07:00
page->mapping = inode->i_mapping;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
page->index = em_start >> PAGE_SHIFT;
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
block: Abstract out bvec iterator Immutable biovecs are going to require an explicit iterator. To implement immutable bvecs, a later patch is going to add a bi_bvec_done member to this struct; for now, this patch effectively just renames things. Signed-off-by: Kent Overstreet <kmo@daterainc.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: "Ed L. Cashin" <ecashin@coraid.com> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Lars Ellenberg <drbd-dev@lists.linbit.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Matthew Wilcox <willy@linux.intel.com> Cc: Geoff Levand <geoff@infradead.org> Cc: Yehuda Sadeh <yehuda@inktank.com> Cc: Sage Weil <sage@inktank.com> Cc: Alex Elder <elder@inktank.com> Cc: ceph-devel@vger.kernel.org Cc: Joshua Morris <josh.h.morris@us.ibm.com> Cc: Philip Kelleher <pjk1939@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Neil Brown <neilb@suse.de> Cc: Alasdair Kergon <agk@redhat.com> Cc: Mike Snitzer <snitzer@redhat.com> Cc: dm-devel@redhat.com Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: linux390@de.ibm.com Cc: Boaz Harrosh <bharrosh@panasas.com> Cc: Benny Halevy <bhalevy@tonian.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Chris Mason <chris.mason@fusionio.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: Jaegeuk Kim <jaegeuk.kim@samsung.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@kernel.org> Cc: Joern Engel <joern@logfs.org> Cc: Prasad Joshi <prasadjoshi.linux@gmail.com> Cc: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: KONISHI Ryusuke <konishi.ryusuke@lab.ntt.co.jp> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Ben Myers <bpm@sgi.com> Cc: xfs@oss.sgi.com Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Len Brown <len.brown@intel.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Herton Ronaldo Krzesinski <herton.krzesinski@canonical.com> Cc: Ben Hutchings <ben@decadent.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Guo Chao <yan@linux.vnet.ibm.com> Cc: Tejun Heo <tj@kernel.org> Cc: Asai Thambi S P <asamymuthupa@micron.com> Cc: Selvan Mani <smani@micron.com> Cc: Sam Bradshaw <sbradshaw@micron.com> Cc: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Cc: "Roger Pau Monné" <roger.pau@citrix.com> Cc: Jan Beulich <jbeulich@suse.com> Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Cc: Ian Campbell <Ian.Campbell@citrix.com> Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Jiang Liu <jiang.liu@huawei.com> Cc: Nitin Gupta <ngupta@vflare.org> Cc: Jerome Marchand <jmarchand@redhat.com> Cc: Joe Perches <joe@perches.com> Cc: Peng Tao <tao.peng@emc.com> Cc: Andy Adamson <andros@netapp.com> Cc: fanchaoting <fanchaoting@cn.fujitsu.com> Cc: Jie Liu <jeff.liu@oracle.com> Cc: Sunil Mushran <sunil.mushran@gmail.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: Namjae Jeon <namjae.jeon@samsung.com> Cc: Pankaj Kumar <pankaj.km@samsung.com> Cc: Dan Magenheimer <dan.magenheimer@oracle.com> Cc: Mel Gorman <mgorman@suse.de>6
2013-10-12 05:44:27 +07:00
if (comp_bio->bi_iter.bi_size)
ret = tree->ops->merge_bio_hook(page, 0,
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
PAGE_SIZE,
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>
2008-10-30 01:49:59 +07:00
comp_bio, 0);
else
ret = 0;
page->mapping = NULL;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
if (ret || bio_add_page(comp_bio, page, PAGE_SIZE, 0) <
PAGE_SIZE) {
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>
2008-10-30 01:49:59 +07:00
bio_get(comp_bio);
ret = btrfs_bio_wq_end_io(root->fs_info, comp_bio,
BTRFS_WQ_ENDIO_DATA);
BUG_ON(ret); /* -ENOMEM */
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>
2008-10-30 01:49:59 +07:00
/*
* inc the count before we submit the bio so
* we know the end IO handler won't happen before
* we inc the count. Otherwise, the cb might get
* freed before we're done setting it up
*/
atomic_inc(&cb->pending_bios);
if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) {
ret = btrfs_lookup_bio_sums(root, inode,
comp_bio, sums);
BUG_ON(ret); /* -ENOMEM */
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
}
sums += DIV_ROUND_UP(comp_bio->bi_iter.bi_size,
root->sectorsize);
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
ret = btrfs_map_bio(root, comp_bio, mirror_num, 0);
if (ret) {
comp_bio->bi_error = ret;
bio_endio(comp_bio);
}
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>
2008-10-30 01:49:59 +07:00
bio_put(comp_bio);
comp_bio = compressed_bio_alloc(bdev, cur_disk_byte,
GFP_NOFS);
BUG_ON(!comp_bio);
bio_set_op_attrs(comp_bio, REQ_OP_READ, 0);
comp_bio->bi_private = cb;
comp_bio->bi_end_io = end_compressed_bio_read;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
bio_add_page(comp_bio, page, PAGE_SIZE, 0);
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>
2008-10-30 01:49:59 +07:00
}
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
cur_disk_byte += PAGE_SIZE;
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>
2008-10-30 01:49:59 +07:00
}
bio_get(comp_bio);
ret = btrfs_bio_wq_end_io(root->fs_info, comp_bio,
BTRFS_WQ_ENDIO_DATA);
BUG_ON(ret); /* -ENOMEM */
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>
2008-10-30 01:49:59 +07:00
if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) {
ret = btrfs_lookup_bio_sums(root, inode, comp_bio, sums);
BUG_ON(ret); /* -ENOMEM */
}
Btrfs: move data checksumming into a dedicated tree Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-12-09 04:58:54 +07:00
ret = btrfs_map_bio(root, comp_bio, mirror_num, 0);
if (ret) {
comp_bio->bi_error = ret;
bio_endio(comp_bio);
}
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>
2008-10-30 01:49:59 +07:00
bio_put(comp_bio);
return 0;
fail2:
while (faili >= 0) {
__free_page(cb->compressed_pages[faili]);
faili--;
}
kfree(cb->compressed_pages);
fail1:
kfree(cb);
out:
free_extent_map(em);
return 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>
2008-10-30 01:49:59 +07:00
}
static struct {
struct list_head idle_ws;
spinlock_t ws_lock;
/* Number of free workspaces */
int free_ws;
/* Total number of allocated workspaces */
atomic_t total_ws;
/* Waiters for a free workspace */
wait_queue_head_t ws_wait;
} btrfs_comp_ws[BTRFS_COMPRESS_TYPES];
static const struct btrfs_compress_op * const btrfs_compress_op[] = {
&btrfs_zlib_compress,
&btrfs_lzo_compress,
};
void __init btrfs_init_compress(void)
{
int i;
for (i = 0; i < BTRFS_COMPRESS_TYPES; i++) {
struct list_head *workspace;
INIT_LIST_HEAD(&btrfs_comp_ws[i].idle_ws);
spin_lock_init(&btrfs_comp_ws[i].ws_lock);
atomic_set(&btrfs_comp_ws[i].total_ws, 0);
init_waitqueue_head(&btrfs_comp_ws[i].ws_wait);
/*
* Preallocate one workspace for each compression type so
* we can guarantee forward progress in the worst case
*/
workspace = btrfs_compress_op[i]->alloc_workspace();
if (IS_ERR(workspace)) {
pr_warn("BTRFS: cannot preallocate compression workspace, will try later\n");
} else {
atomic_set(&btrfs_comp_ws[i].total_ws, 1);
btrfs_comp_ws[i].free_ws = 1;
list_add(workspace, &btrfs_comp_ws[i].idle_ws);
}
}
}
/*
* This finds an available workspace or allocates a new one.
* If it's not possible to allocate a new one, waits until there's one.
* Preallocation makes a forward progress guarantees and we do not return
* errors.
*/
static struct list_head *find_workspace(int type)
{
struct list_head *workspace;
int cpus = num_online_cpus();
int idx = type - 1;
struct list_head *idle_ws = &btrfs_comp_ws[idx].idle_ws;
spinlock_t *ws_lock = &btrfs_comp_ws[idx].ws_lock;
atomic_t *total_ws = &btrfs_comp_ws[idx].total_ws;
wait_queue_head_t *ws_wait = &btrfs_comp_ws[idx].ws_wait;
int *free_ws = &btrfs_comp_ws[idx].free_ws;
again:
spin_lock(ws_lock);
if (!list_empty(idle_ws)) {
workspace = idle_ws->next;
list_del(workspace);
(*free_ws)--;
spin_unlock(ws_lock);
return workspace;
}
if (atomic_read(total_ws) > cpus) {
DEFINE_WAIT(wait);
spin_unlock(ws_lock);
prepare_to_wait(ws_wait, &wait, TASK_UNINTERRUPTIBLE);
if (atomic_read(total_ws) > cpus && !*free_ws)
schedule();
finish_wait(ws_wait, &wait);
goto again;
}
atomic_inc(total_ws);
spin_unlock(ws_lock);
workspace = btrfs_compress_op[idx]->alloc_workspace();
if (IS_ERR(workspace)) {
atomic_dec(total_ws);
wake_up(ws_wait);
/*
* Do not return the error but go back to waiting. There's a
* workspace preallocated for each type and the compression
* time is bounded so we get to a workspace eventually. This
* makes our caller's life easier.
*
* To prevent silent and low-probability deadlocks (when the
* initial preallocation fails), check if there are any
* workspaces at all.
*/
if (atomic_read(total_ws) == 0) {
static DEFINE_RATELIMIT_STATE(_rs,
/* once per minute */ 60 * HZ,
/* no burst */ 1);
if (__ratelimit(&_rs)) {
pr_warn("BTRFS: no compression workspaces, low memory, retrying\n");
}
}
goto again;
}
return workspace;
}
/*
* put a workspace struct back on the list or free it if we have enough
* idle ones sitting around
*/
static void free_workspace(int type, struct list_head *workspace)
{
int idx = type - 1;
struct list_head *idle_ws = &btrfs_comp_ws[idx].idle_ws;
spinlock_t *ws_lock = &btrfs_comp_ws[idx].ws_lock;
atomic_t *total_ws = &btrfs_comp_ws[idx].total_ws;
wait_queue_head_t *ws_wait = &btrfs_comp_ws[idx].ws_wait;
int *free_ws = &btrfs_comp_ws[idx].free_ws;
spin_lock(ws_lock);
if (*free_ws < num_online_cpus()) {
list_add(workspace, idle_ws);
(*free_ws)++;
spin_unlock(ws_lock);
goto wake;
}
spin_unlock(ws_lock);
btrfs_compress_op[idx]->free_workspace(workspace);
atomic_dec(total_ws);
wake:
/*
* Make sure counter is updated before we wake up waiters.
*/
smp_mb();
if (waitqueue_active(ws_wait))
wake_up(ws_wait);
}
/*
* cleanup function for module exit
*/
static void free_workspaces(void)
{
struct list_head *workspace;
int i;
for (i = 0; i < BTRFS_COMPRESS_TYPES; i++) {
while (!list_empty(&btrfs_comp_ws[i].idle_ws)) {
workspace = btrfs_comp_ws[i].idle_ws.next;
list_del(workspace);
btrfs_compress_op[i]->free_workspace(workspace);
atomic_dec(&btrfs_comp_ws[i].total_ws);
}
}
}
/*
* given an address space and start/len, compress the bytes.
*
* pages are allocated to hold the compressed result and stored
* in 'pages'
*
* out_pages is used to return the number of pages allocated. There
* may be pages allocated even if we return an error
*
* total_in is used to return the number of bytes actually read. It
* may be smaller then len if we had to exit early because we
* ran out of room in the pages array or because we cross the
* max_out threshold.
*
* total_out is used to return the total number of compressed bytes
*
* max_out tells us the max number of bytes that we're allowed to
* stuff into pages
*/
int btrfs_compress_pages(int type, struct address_space *mapping,
u64 start, unsigned long len,
struct page **pages,
unsigned long nr_dest_pages,
unsigned long *out_pages,
unsigned long *total_in,
unsigned long *total_out,
unsigned long max_out)
{
struct list_head *workspace;
int ret;
workspace = find_workspace(type);
ret = btrfs_compress_op[type-1]->compress_pages(workspace, mapping,
start, len, pages,
nr_dest_pages, out_pages,
total_in, total_out,
max_out);
free_workspace(type, workspace);
return ret;
}
/*
* pages_in is an array of pages with compressed data.
*
* disk_start is the starting logical offset of this array in the file
*
* bvec is a bio_vec of pages from the file that we want to decompress into
*
* vcnt is the count of pages in the biovec
*
* srclen is the number of bytes in pages_in
*
* The basic idea is that we have a bio that was created by readpages.
* The pages in the bio are for the uncompressed data, and they may not
* be contiguous. They all correspond to the range of bytes covered by
* the compressed extent.
*/
static int btrfs_decompress_biovec(int type, struct page **pages_in,
u64 disk_start, struct bio_vec *bvec,
int vcnt, size_t srclen)
{
struct list_head *workspace;
int ret;
workspace = find_workspace(type);
ret = btrfs_compress_op[type-1]->decompress_biovec(workspace, pages_in,
disk_start,
bvec, vcnt, srclen);
free_workspace(type, workspace);
return ret;
}
/*
* a less complex decompression routine. Our compressed data fits in a
* single page, and we want to read a single page out of it.
* start_byte tells us the offset into the compressed data we're interested in
*/
int btrfs_decompress(int type, unsigned char *data_in, struct page *dest_page,
unsigned long start_byte, size_t srclen, size_t destlen)
{
struct list_head *workspace;
int ret;
workspace = find_workspace(type);
ret = btrfs_compress_op[type-1]->decompress(workspace, data_in,
dest_page, start_byte,
srclen, destlen);
free_workspace(type, workspace);
return ret;
}
void btrfs_exit_compress(void)
{
free_workspaces();
}
/*
* Copy uncompressed data from working buffer to pages.
*
* buf_start is the byte offset we're of the start of our workspace buffer.
*
* total_out is the last byte of the buffer
*/
int btrfs_decompress_buf2page(char *buf, unsigned long buf_start,
unsigned long total_out, u64 disk_start,
struct bio_vec *bvec, int vcnt,
unsigned long *pg_index,
unsigned long *pg_offset)
{
unsigned long buf_offset;
unsigned long current_buf_start;
unsigned long start_byte;
unsigned long working_bytes = total_out - buf_start;
unsigned long bytes;
char *kaddr;
struct page *page_out = bvec[*pg_index].bv_page;
/*
* start byte is the first byte of the page we're currently
* copying into relative to the start of the compressed data.
*/
start_byte = page_offset(page_out) - disk_start;
/* we haven't yet hit data corresponding to this page */
if (total_out <= start_byte)
return 1;
/*
* the start of the data we care about is offset into
* the middle of our working buffer
*/
if (total_out > start_byte && buf_start < start_byte) {
buf_offset = start_byte - buf_start;
working_bytes -= buf_offset;
} else {
buf_offset = 0;
}
current_buf_start = buf_start;
/* copy bytes from the working buffer into the pages */
while (working_bytes > 0) {
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
bytes = min(PAGE_SIZE - *pg_offset,
PAGE_SIZE - buf_offset);
bytes = min(bytes, working_bytes);
kaddr = kmap_atomic(page_out);
memcpy(kaddr + *pg_offset, buf + buf_offset, bytes);
kunmap_atomic(kaddr);
flush_dcache_page(page_out);
*pg_offset += bytes;
buf_offset += bytes;
working_bytes -= bytes;
current_buf_start += bytes;
/* check if we need to pick another page */
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 19:29:47 +07:00
if (*pg_offset == PAGE_SIZE) {
(*pg_index)++;
if (*pg_index >= vcnt)
return 0;
page_out = bvec[*pg_index].bv_page;
*pg_offset = 0;
start_byte = page_offset(page_out) - disk_start;
/*
* make sure our new page is covered by this
* working buffer
*/
if (total_out <= start_byte)
return 1;
/*
* the next page in the biovec might not be adjacent
* to the last page, but it might still be found
* inside this working buffer. bump our offset pointer
*/
if (total_out > start_byte &&
current_buf_start < start_byte) {
buf_offset = start_byte - buf_start;
working_bytes = total_out - start_byte;
current_buf_start = buf_start + buf_offset;
}
}
}
return 1;
}
/*
* When uncompressing data, we need to make sure and zero any parts of
* the biovec that were not filled in by the decompression code. pg_index
* and pg_offset indicate the last page and the last offset of that page
* that have been filled in. This will zero everything remaining in the
* biovec.
*/
void btrfs_clear_biovec_end(struct bio_vec *bvec, int vcnt,
unsigned long pg_index,
unsigned long pg_offset)
{
while (pg_index < vcnt) {
struct page *page = bvec[pg_index].bv_page;
unsigned long off = bvec[pg_index].bv_offset;
unsigned long len = bvec[pg_index].bv_len;
if (pg_offset < off)
pg_offset = off;
if (pg_offset < off + len) {
unsigned long bytes = off + len - pg_offset;
char *kaddr;
kaddr = kmap_atomic(page);
memset(kaddr + pg_offset, 0, bytes);
kunmap_atomic(kaddr);
}
pg_index++;
pg_offset = 0;
}
}