linux.git/fs/btrfs/locking.c, branch v4.0 Linux kernel source tree btrfs: fix lockups from btrfs_clear_path_blocking 2014-11-19T18:34:35+00:00 Chris Mason clm@fb.com 2014-11-19T18:25:09+00:00 f82c458a2c3ffb94b431fc6ad791a79df1b3713e The fair reader/writer locks mean that btrfs_clear_path_blocking needs to strictly follow lock ordering rules even when we already have blocking locks on a given path. Before we can clear a blocking lock on the path, we need to make sure all of the locks have been converted to blocking. This will remove lock inversions against anyone spinning in write_lock() against the buffers we're trying to get read locks on. These inversions didn't exist before the fair read/writer locks, but now we need to be more careful. We papered over this deadlock in the past by changing btrfs_try_read_lock() to be a true trylock against both the spinlock and the blocking lock. This was slower, and not sufficient to fix all the deadlocks. This patch adds a btrfs_tree_read_lock_atomic(), which basically means get the spinlock but trylock on the blocking lock. Signed-off-by: Chris Mason <clm@fb.com> Signed-off-by: Josef Bacik <jbacik@fb.com> Reported-by: Patrick Schmid <schmid@phys.ethz.ch> cc: stable@vger.kernel.org #v3.15+ 
The fair reader/writer locks mean that btrfs_clear_path_blocking needs
to strictly follow lock ordering rules even when we already have
blocking locks on a given path.

Before we can clear a blocking lock on the path, we need to make sure
all of the locks have been converted to blocking.  This will remove lock
inversions against anyone spinning in write_lock() against the buffers
we're trying to get read locks on.  These inversions didn't exist before
the fair read/writer locks, but now we need to be more careful.

We papered over this deadlock in the past by changing
btrfs_try_read_lock() to be a true trylock against both the spinlock and
the blocking lock.  This was slower, and not sufficient to fix all the
deadlocks.  This patch adds a btrfs_tree_read_lock_atomic(), which
basically means get the spinlock but trylock on the blocking lock.

Signed-off-by: Chris Mason <clm@fb.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reported-by: Patrick Schmid <schmid@phys.ethz.ch>
cc: stable@vger.kernel.org #v3.15+
Btrfs: fix deadlocks with trylock on tree nodes 2014-06-19T21:19:55+00:00 Chris Mason clm@fb.com 2014-06-19T21:16:52+00:00 ea4ebde02e08558b020c4b61bb9a4c0fcf63028e The Btrfs tree trylock function is poorly named. It always takes the spinlock and backs off if the blocking lock is held. This can lead to surprising lockups because people expect it to really be a trylock. This commit makes it a pure trylock, both for the spinlock and the blocking lock. It also reworks the nested lock handling slightly to avoid taking the read lock while a spinning write lock might be held. Signed-off-by: Chris Mason <clm@fb.com> 
The Btrfs tree trylock function is poorly named.  It always takes
the spinlock and backs off if the blocking lock is held.  This
can lead to surprising lockups because people expect it to really be a
trylock.

This commit makes it a pure trylock, both for the spinlock and the
blocking lock.  It also reworks the nested lock handling slightly to
avoid taking the read lock while a spinning write lock might be held.

Signed-off-by: Chris Mason <clm@fb.com>
btrfs: make static code static & remove dead code 2013-05-06T19:55:23+00:00 Eric Sandeen sandeen@redhat.com 2013-04-25T20:41:01+00:00 48a3b6366f6913683563d934eb16fea67dead9c1 Big patch, but all it does is add statics to functions which are in fact static, then remove the associated dead-code fallout. removed functions: btrfs_iref_to_path() __btrfs_lookup_delayed_deletion_item() __btrfs_search_delayed_insertion_item() __btrfs_search_delayed_deletion_item() find_eb_for_page() btrfs_find_block_group() range_straddles_pages() extent_range_uptodate() btrfs_file_extent_length() btrfs_scrub_cancel_devid() btrfs_start_transaction_lflush() btrfs_print_tree() is left because it is used for debugging. btrfs_start_transaction_lflush() and btrfs_reada_detach() are left for symmetry. ulist.c functions are left, another patch will take care of those. Signed-off-by: Eric Sandeen <sandeen@redhat.com> Signed-off-by: Josef Bacik <jbacik@fusionio.com> 
Big patch, but all it does is add statics to functions which
are in fact static, then remove the associated dead-code fallout.

removed functions:

btrfs_iref_to_path()
__btrfs_lookup_delayed_deletion_item()
__btrfs_search_delayed_insertion_item()
__btrfs_search_delayed_deletion_item()
find_eb_for_page()
btrfs_find_block_group()
range_straddles_pages()
extent_range_uptodate()
btrfs_file_extent_length()
btrfs_scrub_cancel_devid()
btrfs_start_transaction_lflush()

btrfs_print_tree() is left because it is used for debugging.
btrfs_start_transaction_lflush() and btrfs_reada_detach() are
left for symmetry.

ulist.c functions are left, another patch will take care of those.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Btrfs: save us a read_lock 2013-02-20T14:37:17+00:00 Liu Bo bo.li.liu@oracle.com 2012-12-27T09:01:22+00:00 39f9d028c9906cc7b625df84442205f6bcc84477 This does not change the logic of code, but can save us a read_lock. Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: Josef Bacik <jbacik@fusionio.com> 
This does not change the logic of code, but can save us a read_lock.

Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Btrfs: fix a misplaced address operator in a condition 2012-08-28T20:53:23+00:00 Stefan Behrens sbehrens@giantdisaster.de 2012-07-26T09:40:35+00:00 aa2ffd06168e25689e0eb9662bf4595ba2bbac14 This should obviously not be "if (&flag)" but "if (flag)". Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de> 
This should obviously not be "if (&flag)" but "if (flag)".

Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
Btrfs: reduce calls to wake_up on uncontended locks 2012-07-23T19:36:18+00:00 Chris Mason chris.mason@fusionio.com 2012-07-23T19:25:05+00:00 cbea5ac1ee03197354bd38caad3fcb798f185181 The btrfs locks were unconditionally calling wake_up as the locks were released. This lead to extra thrashing on the waitqueue, especially for locks that were dominated by readers. Signed-off-by: Chris Mason <chris.mason@fusionio.com> 
The btrfs locks were unconditionally calling wake_up as the
locks were released.  This lead to extra thrashing on the waitqueue,
especially for locks that were dominated by readers.

Signed-off-by: Chris Mason <chris.mason@fusionio.com>
btrfs: return void in functions without error conditions 2012-03-22T00:45:34+00:00 Jeff Mahoney jeffm@suse.com 2012-03-01T13:56:26+00:00 143bede527b054a271053f41bfaca2b57baa9408 Signed-off-by: Jeff Mahoney <jeffm@suse.com> 
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Btrfs: add nested locking mode for paths 2012-01-04T15:12:29+00:00 Arne Jansen sensille@gmx.net 2011-09-13T08:55:48+00:00 5b25f70f4200766355cdabda604e131d2fb6010d This patch adds the possibilty to read-lock an extent even if it is already write-locked from the same thread. btrfs_find_all_roots() needs this capability. Signed-off-by: Arne Jansen <sensille@gmx.net> Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net> 
This patch adds the possibilty to read-lock an extent even if it is already
write-locked from the same thread. btrfs_find_all_roots() needs this
capability.

Signed-off-by: Arne Jansen <sensille@gmx.net>
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
Btrfs: switch the btrfs tree locks to reader/writer 2011-07-27T16:46:46+00:00 Chris Mason chris.mason@oracle.com 2011-07-16T19:23:14+00:00 bd681513fa6f2ff29aa391f01e413a2d1c59fd77 The btrfs metadata btree is the source of significant lock contention, especially in the root node. This commit changes our locking to use a reader/writer lock. The lock is built on top of rw spinlocks, and it extends the lock tracking to remember if we have a read lock or a write lock when we go to blocking. Atomics count the number of blocking readers or writers at any given time. It removes all of the adaptive spinning from the old code and uses only the spinning/blocking hints inside of btrfs to decide when it should continue spinning. In read heavy workloads this is dramatically faster. In write heavy workloads we're still faster because of less contention on the root node lock. We suffer slightly in dbench because we schedule more often during write locks, but all other benchmarks so far are improved. Signed-off-by: Chris Mason <chris.mason@oracle.com> 
The btrfs metadata btree is the source of significant
lock contention, especially in the root node.   This
commit changes our locking to use a reader/writer
lock.

The lock is built on top of rw spinlocks, and it
extends the lock tracking to remember if we have a
read lock or a write lock when we go to blocking.  Atomics
count the number of blocking readers or writers at any
given time.

It removes all of the adaptive spinning from the old code
and uses only the spinning/blocking hints inside of btrfs
to decide when it should continue spinning.

In read heavy workloads this is dramatically faster.  In write
heavy workloads we're still faster because of less contention
on the root node lock.

We suffer slightly in dbench because we schedule more often
during write locks, but all other benchmarks so far are improved.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
btrfs: remove all unused functions 2011-05-06T10:34:03+00:00 David Sterba dsterba@suse.cz 2011-05-05T10:44:41+00:00 f2a97a9dbd86eb1ef956bdf20e05c507b32beb96 Remove static and global declarations and/or definitions. Reduces size of btrfs.ko by ~3.4kB. text data bss dec hex filename 402081 7464 200 409745 64091 btrfs.ko.base 398620 7144 200 405964 631cc btrfs.ko.remove-all Signed-off-by: David Sterba <dsterba@suse.cz> 
Remove static and global declarations and/or definitions. Reduces size
of btrfs.ko by ~3.4kB.

  text    data     bss     dec     hex filename
402081    7464     200  409745   64091 btrfs.ko.base
398620    7144     200  405964   631cc btrfs.ko.remove-all

Signed-off-by: David Sterba <dsterba@suse.cz>