linux_dsm_epyc7002/fs/btrfs/extent_io.h

382 lines
14 KiB
C
Raw Normal View History

#ifndef __EXTENTIO__
#define __EXTENTIO__
#include <linux/rbtree.h>
/* bits for the extent state */
#define EXTENT_DIRTY 1
#define EXTENT_WRITEBACK (1 << 1)
#define EXTENT_UPTODATE (1 << 2)
#define EXTENT_LOCKED (1 << 3)
#define EXTENT_NEW (1 << 4)
#define EXTENT_DELALLOC (1 << 5)
#define EXTENT_DEFRAG (1 << 6)
#define EXTENT_BOUNDARY (1 << 9)
#define EXTENT_NODATASUM (1 << 10)
#define EXTENT_DO_ACCOUNTING (1 << 11)
#define EXTENT_FIRST_DELALLOC (1 << 12)
#define EXTENT_NEED_WAIT (1 << 13)
#define EXTENT_DAMAGED (1 << 14)
#define EXTENT_NORESERVE (1 << 15)
#define EXTENT_IOBITS (EXTENT_LOCKED | EXTENT_WRITEBACK)
#define EXTENT_CTLBITS (EXTENT_DO_ACCOUNTING | EXTENT_FIRST_DELALLOC)
/*
* flags for bio submission. The high bits indicate the compression
* type for this 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
#define EXTENT_BIO_COMPRESSED 1
#define EXTENT_BIO_TREE_LOG 2
#define EXTENT_BIO_PARENT_LOCKED 4
#define EXTENT_BIO_FLAG_SHIFT 16
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: Change btree locking to use explicit blocking points Most of the btrfs metadata operations can be protected by a spinlock, but some operations still need to schedule. So far, btrfs has been using a mutex along with a trylock loop, most of the time it is able to avoid going for the full mutex, so the trylock loop is a big performance gain. This commit is step one for getting rid of the blocking locks entirely. btrfs_tree_lock takes a spinlock, and the code explicitly switches to a blocking lock when it starts an operation that can schedule. We'll be able get rid of the blocking locks in smaller pieces over time. Tracing allows us to find the most common cause of blocking, so we can start with the hot spots first. The basic idea is: btrfs_tree_lock() returns with the spin lock held btrfs_set_lock_blocking() sets the EXTENT_BUFFER_BLOCKING bit in the extent buffer flags, and then drops the spin lock. The buffer is still considered locked by all of the btrfs code. If btrfs_tree_lock gets the spinlock but finds the blocking bit set, it drops the spin lock and waits on a wait queue for the blocking bit to go away. Much of the code that needs to set the blocking bit finishes without actually blocking a good percentage of the time. So, an adaptive spin is still used against the blocking bit to avoid very high context switch rates. btrfs_clear_lock_blocking() clears the blocking bit and returns with the spinlock held again. btrfs_tree_unlock() can be called on either blocking or spinning locks, it does the right thing based on the blocking bit. ctree.c has a helper function to set/clear all the locked buffers in a path as blocking. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-02-04 21:25:08 +07:00
/* these are bit numbers for test/set bit */
#define EXTENT_BUFFER_UPTODATE 0
#define EXTENT_BUFFER_DIRTY 2
#define EXTENT_BUFFER_CORRUPT 3
#define EXTENT_BUFFER_READAHEAD 4 /* this got triggered by readahead */
#define EXTENT_BUFFER_TREE_REF 5
#define EXTENT_BUFFER_STALE 6
#define EXTENT_BUFFER_WRITEBACK 7
Btrfs: be aware of btree inode write errors to avoid fs corruption While we have a transaction ongoing, the VM might decide at any time to call btree_inode->i_mapping->a_ops->writepages(), which will start writeback of dirty pages belonging to btree nodes/leafs. This call might return an error or the writeback might finish with an error before we attempt to commit the running transaction. If this happens, we might have no way of knowing that such error happened when we are committing the transaction - because the pages might no longer be marked dirty nor tagged for writeback (if a subsequent modification to the extent buffer didn't happen before the transaction commit) which makes filemap_fdata[write|wait]_range unable to find such pages (even if they're marked with SetPageError). So if this happens we must abort the transaction, otherwise we commit a super block with btree roots that point to btree nodes/leafs whose content on disk is invalid - either garbage or the content of some node/leaf from a past generation that got cowed or deleted and is no longer valid (for this later case we end up getting error messages like "parent transid verify failed on 10826481664 wanted 25748 found 29562" when reading btree nodes/leafs from disk). Note that setting and checking AS_EIO/AS_ENOSPC in the btree inode's i_mapping would not be enough because we need to distinguish between log tree extents (not fatal) vs non-log tree extents (fatal) and because the next call to filemap_fdatawait_range() will catch and clear such errors in the mapping - and that call might be from a log sync and not from a transaction commit, which means we would not know about the error at transaction commit time. Also, checking for the eb flag EXTENT_BUFFER_IOERR at transaction commit time isn't done and would not be completely reliable, as the eb might be removed from memory and read back when trying to get it, which clears that flag right before reading the eb's pages from disk, making us not know about the previous write error. Using the new 3 flags for the btree inode also makes us achieve the goal of AS_EIO/AS_ENOSPC when writepages() returns success, started writeback for all dirty pages and before filemap_fdatawait_range() is called, the writeback for all dirty pages had already finished with errors - because we were not using AS_EIO/AS_ENOSPC, filemap_fdatawait_range() would return success, as it could not know that writeback errors happened (the pages were no longer tagged for writeback). Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Chris Mason <clm@fb.com>
2014-09-26 18:25:56 +07:00
#define EXTENT_BUFFER_READ_ERR 8 /* read IO error */
#define EXTENT_BUFFER_DUMMY 9
#define EXTENT_BUFFER_IN_TREE 10
Btrfs: be aware of btree inode write errors to avoid fs corruption While we have a transaction ongoing, the VM might decide at any time to call btree_inode->i_mapping->a_ops->writepages(), which will start writeback of dirty pages belonging to btree nodes/leafs. This call might return an error or the writeback might finish with an error before we attempt to commit the running transaction. If this happens, we might have no way of knowing that such error happened when we are committing the transaction - because the pages might no longer be marked dirty nor tagged for writeback (if a subsequent modification to the extent buffer didn't happen before the transaction commit) which makes filemap_fdata[write|wait]_range unable to find such pages (even if they're marked with SetPageError). So if this happens we must abort the transaction, otherwise we commit a super block with btree roots that point to btree nodes/leafs whose content on disk is invalid - either garbage or the content of some node/leaf from a past generation that got cowed or deleted and is no longer valid (for this later case we end up getting error messages like "parent transid verify failed on 10826481664 wanted 25748 found 29562" when reading btree nodes/leafs from disk). Note that setting and checking AS_EIO/AS_ENOSPC in the btree inode's i_mapping would not be enough because we need to distinguish between log tree extents (not fatal) vs non-log tree extents (fatal) and because the next call to filemap_fdatawait_range() will catch and clear such errors in the mapping - and that call might be from a log sync and not from a transaction commit, which means we would not know about the error at transaction commit time. Also, checking for the eb flag EXTENT_BUFFER_IOERR at transaction commit time isn't done and would not be completely reliable, as the eb might be removed from memory and read back when trying to get it, which clears that flag right before reading the eb's pages from disk, making us not know about the previous write error. Using the new 3 flags for the btree inode also makes us achieve the goal of AS_EIO/AS_ENOSPC when writepages() returns success, started writeback for all dirty pages and before filemap_fdatawait_range() is called, the writeback for all dirty pages had already finished with errors - because we were not using AS_EIO/AS_ENOSPC, filemap_fdatawait_range() would return success, as it could not know that writeback errors happened (the pages were no longer tagged for writeback). Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Chris Mason <clm@fb.com>
2014-09-26 18:25:56 +07:00
#define EXTENT_BUFFER_WRITE_ERR 11 /* write IO error */
Btrfs: Change btree locking to use explicit blocking points Most of the btrfs metadata operations can be protected by a spinlock, but some operations still need to schedule. So far, btrfs has been using a mutex along with a trylock loop, most of the time it is able to avoid going for the full mutex, so the trylock loop is a big performance gain. This commit is step one for getting rid of the blocking locks entirely. btrfs_tree_lock takes a spinlock, and the code explicitly switches to a blocking lock when it starts an operation that can schedule. We'll be able get rid of the blocking locks in smaller pieces over time. Tracing allows us to find the most common cause of blocking, so we can start with the hot spots first. The basic idea is: btrfs_tree_lock() returns with the spin lock held btrfs_set_lock_blocking() sets the EXTENT_BUFFER_BLOCKING bit in the extent buffer flags, and then drops the spin lock. The buffer is still considered locked by all of the btrfs code. If btrfs_tree_lock gets the spinlock but finds the blocking bit set, it drops the spin lock and waits on a wait queue for the blocking bit to go away. Much of the code that needs to set the blocking bit finishes without actually blocking a good percentage of the time. So, an adaptive spin is still used against the blocking bit to avoid very high context switch rates. btrfs_clear_lock_blocking() clears the blocking bit and returns with the spinlock held again. btrfs_tree_unlock() can be called on either blocking or spinning locks, it does the right thing based on the blocking bit. ctree.c has a helper function to set/clear all the locked buffers in a path as blocking. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-02-04 21:25:08 +07:00
/* these are flags for extent_clear_unlock_delalloc */
#define PAGE_UNLOCK (1 << 0)
#define PAGE_CLEAR_DIRTY (1 << 1)
#define PAGE_SET_WRITEBACK (1 << 2)
#define PAGE_END_WRITEBACK (1 << 3)
#define PAGE_SET_PRIVATE2 (1 << 4)
#define PAGE_SET_ERROR (1 << 5)
/*
* page->private values. Every page that is controlled by the extent
* map has page->private set to one.
*/
#define EXTENT_PAGE_PRIVATE 1
struct extent_state;
struct btrfs_root;
struct btrfs_io_bio;
typedef int (extent_submit_bio_hook_t)(struct inode *inode, int rw,
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 bio *bio, int mirror_num,
unsigned long bio_flags, u64 bio_offset);
struct extent_io_ops {
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
int (*fill_delalloc)(struct inode *inode, struct page *locked_page,
u64 start, u64 end, int *page_started,
unsigned long *nr_written);
int (*writepage_start_hook)(struct page *page, u64 start, u64 end);
int (*writepage_io_hook)(struct page *page, u64 start, u64 end);
extent_submit_bio_hook_t *submit_bio_hook;
int (*merge_bio_hook)(int rw, struct page *page, unsigned long offset,
Btrfs: Add zlib compression support This is a large change for adding compression on reading and writing, both for inline and regular extents. It does some fairly large surgery to the writeback paths. Compression is off by default and enabled by mount -o compress. Even when the -o compress mount option is not used, it is possible to read compressed extents off the disk. If compression for a given set of pages fails to make them smaller, the file is flagged to avoid future compression attempts later. * While finding delalloc extents, the pages are locked before being sent down to the delalloc handler. This allows the delalloc handler to do complex things such as cleaning the pages, marking them writeback and starting IO on their behalf. * Inline extents are inserted at delalloc time now. This allows us to compress the data before inserting the inline extent, and it allows us to insert an inline extent that spans multiple pages. * All of the in-memory extent representations (extent_map.c, ordered-data.c etc) are changed to record both an in-memory size and an on disk size, as well as a flag for compression. From a disk format point of view, the extent pointers in the file are changed to record the on disk size of a given extent and some encoding flags. Space in the disk format is allocated for compression encoding, as well as encryption and a generic 'other' field. Neither the encryption or the 'other' field are currently used. In order to limit the amount of data read for a single random read in the file, the size of a compressed extent is limited to 128k. This is a software only limit, the disk format supports u64 sized compressed extents. In order to limit the ram consumed while processing extents, the uncompressed size of a compressed extent is limited to 256k. This is a software only limit and will be subject to tuning later. Checksumming is still done on compressed extents, and it is done on the uncompressed version of the data. This way additional encodings can be layered on without having to figure out which encoding to checksum. Compression happens at delalloc time, which is basically singled threaded because it is usually done by a single pdflush thread. This makes it tricky to spread the compression load across all the cpus on the box. We'll have to look at parallel pdflush walks of dirty inodes at a later time. Decompression is hooked into readpages and it does spread across CPUs nicely. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-10-30 01:49:59 +07:00
size_t size, struct bio *bio,
unsigned long bio_flags);
int (*readpage_io_failed_hook)(struct page *page, int failed_mirror);
int (*readpage_end_io_hook)(struct btrfs_io_bio *io_bio, u64 phy_offset,
struct page *page, u64 start, u64 end,
int mirror);
int (*writepage_end_io_hook)(struct page *page, u64 start, u64 end,
struct extent_state *state, int uptodate);
void (*set_bit_hook)(struct inode *inode, struct extent_state *state,
unsigned long *bits);
void (*clear_bit_hook)(struct inode *inode, struct extent_state *state,
unsigned long *bits);
void (*merge_extent_hook)(struct inode *inode,
struct extent_state *new,
struct extent_state *other);
void (*split_extent_hook)(struct inode *inode,
struct extent_state *orig, u64 split);
};
struct extent_io_tree {
struct rb_root state;
struct address_space *mapping;
u64 dirty_bytes;
int track_uptodate;
spinlock_t lock;
struct extent_io_ops *ops;
};
struct extent_state {
u64 start;
u64 end; /* inclusive */
struct rb_node rb_node;
Btrfs: proper -ENOSPC handling At the start of a transaction we do a btrfs_reserve_metadata_space() and specify how many items we plan on modifying. Then once we've done our modifications and such, just call btrfs_unreserve_metadata_space() for the same number of items we reserved. For keeping track of metadata needed for data I've had to add an extent_io op for when we merge extents. This lets us track space properly when we are doing sequential writes, so we don't end up reserving way more metadata space than what we need. The only place where the metadata space accounting is not done is in the relocation code. This is because Yan is going to be reworking that code in the near future, so running btrfs-vol -b could still possibly result in a ENOSPC related panic. This patch also turns off the metadata_ratio stuff in order to allow users to more efficiently use their disk space. This patch makes it so we track how much metadata we need for an inode's delayed allocation extents by tracking how many extents are currently waiting for allocation. It introduces two new callbacks for the extent_io tree's, merge_extent_hook and split_extent_hook. These help us keep track of when we merge delalloc extents together and split them up. Reservations are handled prior to any actually dirty'ing occurs, and then we unreserve after we dirty. btrfs_unreserve_metadata_for_delalloc() will make the appropriate unreservations as needed based on the number of reservations we currently have and the number of extents we currently have. Doing the reservation outside of doing any of the actual dirty'ing lets us do things like filemap_flush() the inode to try and force delalloc to happen, or as a last resort actually start allocation on all delalloc inodes in the fs. This has survived dbench, fs_mark and an fsx torture test. Signed-off-by: Josef Bacik <jbacik@redhat.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-09-12 03:12:44 +07:00
/* ADD NEW ELEMENTS AFTER THIS */
wait_queue_head_t wq;
atomic_t refs;
unsigned long state;
/* for use by the FS */
u64 private;
#ifdef CONFIG_BTRFS_DEBUG
struct list_head leak_list;
#endif
};
#define INLINE_EXTENT_BUFFER_PAGES 16
#define MAX_INLINE_EXTENT_BUFFER_SIZE (INLINE_EXTENT_BUFFER_PAGES * PAGE_CACHE_SIZE)
struct extent_buffer {
u64 start;
unsigned long len;
Btrfs: Change btree locking to use explicit blocking points Most of the btrfs metadata operations can be protected by a spinlock, but some operations still need to schedule. So far, btrfs has been using a mutex along with a trylock loop, most of the time it is able to avoid going for the full mutex, so the trylock loop is a big performance gain. This commit is step one for getting rid of the blocking locks entirely. btrfs_tree_lock takes a spinlock, and the code explicitly switches to a blocking lock when it starts an operation that can schedule. We'll be able get rid of the blocking locks in smaller pieces over time. Tracing allows us to find the most common cause of blocking, so we can start with the hot spots first. The basic idea is: btrfs_tree_lock() returns with the spin lock held btrfs_set_lock_blocking() sets the EXTENT_BUFFER_BLOCKING bit in the extent buffer flags, and then drops the spin lock. The buffer is still considered locked by all of the btrfs code. If btrfs_tree_lock gets the spinlock but finds the blocking bit set, it drops the spin lock and waits on a wait queue for the blocking bit to go away. Much of the code that needs to set the blocking bit finishes without actually blocking a good percentage of the time. So, an adaptive spin is still used against the blocking bit to avoid very high context switch rates. btrfs_clear_lock_blocking() clears the blocking bit and returns with the spinlock held again. btrfs_tree_unlock() can be called on either blocking or spinning locks, it does the right thing based on the blocking bit. ctree.c has a helper function to set/clear all the locked buffers in a path as blocking. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-02-04 21:25:08 +07:00
unsigned long bflags;
struct btrfs_fs_info *fs_info;
spinlock_t refs_lock;
atomic_t refs;
atomic_t io_pages;
int read_mirror;
struct rcu_head rcu_head;
pid_t lock_owner;
Btrfs: Change btree locking to use explicit blocking points Most of the btrfs metadata operations can be protected by a spinlock, but some operations still need to schedule. So far, btrfs has been using a mutex along with a trylock loop, most of the time it is able to avoid going for the full mutex, so the trylock loop is a big performance gain. This commit is step one for getting rid of the blocking locks entirely. btrfs_tree_lock takes a spinlock, and the code explicitly switches to a blocking lock when it starts an operation that can schedule. We'll be able get rid of the blocking locks in smaller pieces over time. Tracing allows us to find the most common cause of blocking, so we can start with the hot spots first. The basic idea is: btrfs_tree_lock() returns with the spin lock held btrfs_set_lock_blocking() sets the EXTENT_BUFFER_BLOCKING bit in the extent buffer flags, and then drops the spin lock. The buffer is still considered locked by all of the btrfs code. If btrfs_tree_lock gets the spinlock but finds the blocking bit set, it drops the spin lock and waits on a wait queue for the blocking bit to go away. Much of the code that needs to set the blocking bit finishes without actually blocking a good percentage of the time. So, an adaptive spin is still used against the blocking bit to avoid very high context switch rates. btrfs_clear_lock_blocking() clears the blocking bit and returns with the spinlock held again. btrfs_tree_unlock() can be called on either blocking or spinning locks, it does the right thing based on the blocking bit. ctree.c has a helper function to set/clear all the locked buffers in a path as blocking. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-02-04 21:25:08 +07:00
/* count of read lock holders on the extent buffer */
atomic_t write_locks;
atomic_t read_locks;
atomic_t blocking_writers;
atomic_t blocking_readers;
atomic_t spinning_readers;
atomic_t spinning_writers;
Btrfs: be aware of btree inode write errors to avoid fs corruption While we have a transaction ongoing, the VM might decide at any time to call btree_inode->i_mapping->a_ops->writepages(), which will start writeback of dirty pages belonging to btree nodes/leafs. This call might return an error or the writeback might finish with an error before we attempt to commit the running transaction. If this happens, we might have no way of knowing that such error happened when we are committing the transaction - because the pages might no longer be marked dirty nor tagged for writeback (if a subsequent modification to the extent buffer didn't happen before the transaction commit) which makes filemap_fdata[write|wait]_range unable to find such pages (even if they're marked with SetPageError). So if this happens we must abort the transaction, otherwise we commit a super block with btree roots that point to btree nodes/leafs whose content on disk is invalid - either garbage or the content of some node/leaf from a past generation that got cowed or deleted and is no longer valid (for this later case we end up getting error messages like "parent transid verify failed on 10826481664 wanted 25748 found 29562" when reading btree nodes/leafs from disk). Note that setting and checking AS_EIO/AS_ENOSPC in the btree inode's i_mapping would not be enough because we need to distinguish between log tree extents (not fatal) vs non-log tree extents (fatal) and because the next call to filemap_fdatawait_range() will catch and clear such errors in the mapping - and that call might be from a log sync and not from a transaction commit, which means we would not know about the error at transaction commit time. Also, checking for the eb flag EXTENT_BUFFER_IOERR at transaction commit time isn't done and would not be completely reliable, as the eb might be removed from memory and read back when trying to get it, which clears that flag right before reading the eb's pages from disk, making us not know about the previous write error. Using the new 3 flags for the btree inode also makes us achieve the goal of AS_EIO/AS_ENOSPC when writepages() returns success, started writeback for all dirty pages and before filemap_fdatawait_range() is called, the writeback for all dirty pages had already finished with errors - because we were not using AS_EIO/AS_ENOSPC, filemap_fdatawait_range() would return success, as it could not know that writeback errors happened (the pages were no longer tagged for writeback). Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Chris Mason <clm@fb.com>
2014-09-26 18:25:56 +07:00
short lock_nested;
/* >= 0 if eb belongs to a log tree, -1 otherwise */
short log_index;
/* protects write locks */
rwlock_t lock;
/* readers use lock_wq while they wait for the write
* lock holders to unlock
*/
wait_queue_head_t write_lock_wq;
Btrfs: Change btree locking to use explicit blocking points Most of the btrfs metadata operations can be protected by a spinlock, but some operations still need to schedule. So far, btrfs has been using a mutex along with a trylock loop, most of the time it is able to avoid going for the full mutex, so the trylock loop is a big performance gain. This commit is step one for getting rid of the blocking locks entirely. btrfs_tree_lock takes a spinlock, and the code explicitly switches to a blocking lock when it starts an operation that can schedule. We'll be able get rid of the blocking locks in smaller pieces over time. Tracing allows us to find the most common cause of blocking, so we can start with the hot spots first. The basic idea is: btrfs_tree_lock() returns with the spin lock held btrfs_set_lock_blocking() sets the EXTENT_BUFFER_BLOCKING bit in the extent buffer flags, and then drops the spin lock. The buffer is still considered locked by all of the btrfs code. If btrfs_tree_lock gets the spinlock but finds the blocking bit set, it drops the spin lock and waits on a wait queue for the blocking bit to go away. Much of the code that needs to set the blocking bit finishes without actually blocking a good percentage of the time. So, an adaptive spin is still used against the blocking bit to avoid very high context switch rates. btrfs_clear_lock_blocking() clears the blocking bit and returns with the spinlock held again. btrfs_tree_unlock() can be called on either blocking or spinning locks, it does the right thing based on the blocking bit. ctree.c has a helper function to set/clear all the locked buffers in a path as blocking. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-02-04 21:25:08 +07:00
/* writers use read_lock_wq while they wait for readers
* to unlock
Btrfs: Change btree locking to use explicit blocking points Most of the btrfs metadata operations can be protected by a spinlock, but some operations still need to schedule. So far, btrfs has been using a mutex along with a trylock loop, most of the time it is able to avoid going for the full mutex, so the trylock loop is a big performance gain. This commit is step one for getting rid of the blocking locks entirely. btrfs_tree_lock takes a spinlock, and the code explicitly switches to a blocking lock when it starts an operation that can schedule. We'll be able get rid of the blocking locks in smaller pieces over time. Tracing allows us to find the most common cause of blocking, so we can start with the hot spots first. The basic idea is: btrfs_tree_lock() returns with the spin lock held btrfs_set_lock_blocking() sets the EXTENT_BUFFER_BLOCKING bit in the extent buffer flags, and then drops the spin lock. The buffer is still considered locked by all of the btrfs code. If btrfs_tree_lock gets the spinlock but finds the blocking bit set, it drops the spin lock and waits on a wait queue for the blocking bit to go away. Much of the code that needs to set the blocking bit finishes without actually blocking a good percentage of the time. So, an adaptive spin is still used against the blocking bit to avoid very high context switch rates. btrfs_clear_lock_blocking() clears the blocking bit and returns with the spinlock held again. btrfs_tree_unlock() can be called on either blocking or spinning locks, it does the right thing based on the blocking bit. ctree.c has a helper function to set/clear all the locked buffers in a path as blocking. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2009-02-04 21:25:08 +07:00
*/
wait_queue_head_t read_lock_wq;
struct page *pages[INLINE_EXTENT_BUFFER_PAGES];
#ifdef CONFIG_BTRFS_DEBUG
struct list_head leak_list;
#endif
};
static inline void extent_set_compress_type(unsigned long *bio_flags,
int compress_type)
{
*bio_flags |= compress_type << EXTENT_BIO_FLAG_SHIFT;
}
static inline int extent_compress_type(unsigned long bio_flags)
{
return bio_flags >> EXTENT_BIO_FLAG_SHIFT;
}
struct extent_map_tree;
typedef struct extent_map *(get_extent_t)(struct inode *inode,
struct page *page,
size_t pg_offset,
u64 start, u64 len,
int create);
void extent_io_tree_init(struct extent_io_tree *tree,
struct address_space *mapping);
int try_release_extent_mapping(struct extent_map_tree *map,
struct extent_io_tree *tree, struct page *page,
gfp_t mask);
int try_release_extent_buffer(struct page *page);
int lock_extent(struct extent_io_tree *tree, u64 start, u64 end);
int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
unsigned long bits, struct extent_state **cached);
int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end);
int unlock_extent_cached(struct extent_io_tree *tree, u64 start, u64 end,
struct extent_state **cached, gfp_t mask);
int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end);
int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
get_extent_t *get_extent, int mirror_num);
int extent_read_full_page_nolock(struct extent_io_tree *tree, struct page *page,
get_extent_t *get_extent, int mirror_num);
int __init extent_io_init(void);
void extent_io_exit(void);
u64 count_range_bits(struct extent_io_tree *tree,
u64 *start, u64 search_end,
u64 max_bytes, unsigned long bits, int contig);
void free_extent_state(struct extent_state *state);
int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
unsigned long bits, int filled,
struct extent_state *cached_state);
int clear_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
unsigned long bits, gfp_t mask);
int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
unsigned long bits, int wake, int delete,
struct extent_state **cached, gfp_t mask);
int set_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
unsigned long bits, gfp_t mask);
int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
unsigned long bits, u64 *failed_start,
struct extent_state **cached_state, gfp_t mask);
int set_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
struct extent_state **cached_state, gfp_t mask);
int clear_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
struct extent_state **cached_state, gfp_t mask);
int set_extent_new(struct extent_io_tree *tree, u64 start, u64 end,
gfp_t mask);
int set_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
gfp_t mask);
int clear_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
gfp_t mask);
int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
unsigned long bits, unsigned long clear_bits,
struct extent_state **cached_state, gfp_t mask);
int set_extent_delalloc(struct extent_io_tree *tree, u64 start, u64 end,
struct extent_state **cached_state, gfp_t mask);
int set_extent_defrag(struct extent_io_tree *tree, u64 start, u64 end,
struct extent_state **cached_state, gfp_t mask);
int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
u64 *start_ret, u64 *end_ret, unsigned long bits,
struct extent_state **cached_state);
int extent_invalidatepage(struct extent_io_tree *tree,
struct page *page, unsigned long offset);
int extent_write_full_page(struct extent_io_tree *tree, struct page *page,
get_extent_t *get_extent,
struct writeback_control *wbc);
int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode,
u64 start, u64 end, get_extent_t *get_extent,
int mode);
int extent_writepages(struct extent_io_tree *tree,
struct address_space *mapping,
get_extent_t *get_extent,
struct writeback_control *wbc);
int btree_write_cache_pages(struct address_space *mapping,
struct writeback_control *wbc);
int extent_readpages(struct extent_io_tree *tree,
struct address_space *mapping,
struct list_head *pages, unsigned nr_pages,
get_extent_t get_extent);
int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
__u64 start, __u64 len, get_extent_t *get_extent);
int get_state_private(struct extent_io_tree *tree, u64 start, u64 *private);
void set_page_extent_mapped(struct page *page);
struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
u64 start, unsigned long len);
struct extent_buffer *alloc_dummy_extent_buffer(u64 start, unsigned long len);
struct extent_buffer *btrfs_clone_extent_buffer(struct extent_buffer *src);
struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info,
u64 start);
void free_extent_buffer(struct extent_buffer *eb);
void free_extent_buffer_stale(struct extent_buffer *eb);
#define WAIT_NONE 0
#define WAIT_COMPLETE 1
#define WAIT_PAGE_LOCK 2
int read_extent_buffer_pages(struct extent_io_tree *tree,
struct extent_buffer *eb, u64 start, int wait,
get_extent_t *get_extent, int mirror_num);
void wait_on_extent_buffer_writeback(struct extent_buffer *eb);
static inline unsigned long num_extent_pages(u64 start, u64 len)
{
return ((start + len + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT) -
(start >> PAGE_CACHE_SHIFT);
}
static inline void extent_buffer_get(struct extent_buffer *eb)
{
atomic_inc(&eb->refs);
}
int memcmp_extent_buffer(struct extent_buffer *eb, const void *ptrv,
unsigned long start,
unsigned long len);
void read_extent_buffer(struct extent_buffer *eb, void *dst,
unsigned long start,
unsigned long len);
int read_extent_buffer_to_user(struct extent_buffer *eb, void __user *dst,
unsigned long start,
unsigned long len);
void write_extent_buffer(struct extent_buffer *eb, const void *src,
unsigned long start, unsigned long len);
void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src,
unsigned long dst_offset, unsigned long src_offset,
unsigned long len);
void memcpy_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
unsigned long src_offset, unsigned long len);
void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
unsigned long src_offset, unsigned long len);
void memset_extent_buffer(struct extent_buffer *eb, char c,
unsigned long start, unsigned long len);
void clear_extent_buffer_dirty(struct extent_buffer *eb);
int set_extent_buffer_dirty(struct extent_buffer *eb);
int set_extent_buffer_uptodate(struct extent_buffer *eb);
int clear_extent_buffer_uptodate(struct extent_buffer *eb);
int extent_buffer_uptodate(struct extent_buffer *eb);
int extent_buffer_under_io(struct extent_buffer *eb);
int map_private_extent_buffer(struct extent_buffer *eb, unsigned long offset,
unsigned long min_len, char **map,
unsigned long *map_start,
unsigned long *map_len);
int extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end);
int extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end);
int extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
struct page *locked_page,
unsigned long bits_to_clear,
unsigned long page_ops);
struct bio *
btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
gfp_t gfp_flags);
struct bio *btrfs_io_bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs);
struct bio *btrfs_bio_clone(struct bio *bio, gfp_t gfp_mask);
struct btrfs_fs_info;
int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
struct page *page, unsigned int pg_offset,
int mirror_num);
int clean_io_failure(struct inode *inode, u64 start, struct page *page,
unsigned int pg_offset);
int end_extent_writepage(struct page *page, int err, u64 start, u64 end);
int repair_eb_io_failure(struct btrfs_root *root, struct extent_buffer *eb,
int mirror_num);
/*
* When IO fails, either with EIO or csum verification fails, we
* try other mirrors that might have a good copy of the data. This
* io_failure_record is used to record state as we go through all the
* mirrors. If another mirror has good data, the page is set up to date
* and things continue. If a good mirror can't be found, the original
* bio end_io callback is called to indicate things have failed.
*/
struct io_failure_record {
struct page *page;
u64 start;
u64 len;
u64 logical;
unsigned long bio_flags;
int this_mirror;
int failed_mirror;
int in_validation;
};
void btrfs_free_io_failure_record(struct inode *inode, u64 start, u64 end);
int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
struct io_failure_record **failrec_ret);
int btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
struct io_failure_record *failrec, int fail_mirror);
struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio,
struct io_failure_record *failrec,
struct page *page, int pg_offset, int icsum,
bio_end_io_t *endio_func, void *data);
int free_io_failure(struct inode *inode, struct io_failure_record *rec);
#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
noinline u64 find_lock_delalloc_range(struct inode *inode,
struct extent_io_tree *tree,
struct page *locked_page, u64 *start,
u64 *end, u64 max_bytes);
#endif
struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
u64 start, unsigned long len);
#endif