kernel/git/stable/linux.git/fs/btrfs/locking.h, branch linux-5.11.y

btrfs: remove the recurse parameter from __btrfs_tree_read_lock

2020-12-08T14:54:09+00:00

It is completely unused now, remove it. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba

btrfs: merge back btrfs_read_lock_root_node helpers

2020-12-08T14:54:09+00:00

We no longer have recursive locking and there's no need for separate helpers that allowed the transition to rwsem with minimal code changes. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba

btrfs: locking: remove all the blocking helpers

2020-12-08T14:54:01+00:00

Now that we're using a rw_semaphore we no longer need to indicate if a lock is blocking or not, nor do we need to flip the entire path from blocking to spinning. Remove these helpers and all the places they are called. Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba

btrfs: switch extent buffer tree lock to rw_semaphore

2020-12-08T14:53:43+00:00

Historically we've implemented our own locking because we wanted to be able to selectively spin or sleep based on what we were doing in the tree. For instance, if all of our nodes were in cache then there's rarely a reason to need to sleep waiting for node locks, as they'll likely become available soon. At the time this code was written the rw_semaphore didn't do adaptive spinning, and thus was orders of magnitude slower than our home grown locking. However now the opposite is the case. There are a few problems with how we implement blocking locks, namely that we use a normal waitqueue and simply wake everybody up in reverse sleep order. This leads to some suboptimal performance behavior, and a lot of context switches in highly contended cases. The rw_semaphores actually do this properly, and also have adaptive spinning that works relatively well. The locking code is also a bit of a bear to understand, and we lose the benefit of lockdep for the most part because the blocking states of the lock are simply ad-hoc and not mapped into lockdep. So rework the locking code to drop all of this custom locking stuff, and simply use a rw_semaphore for everything. This makes the locking much simpler for everything, as we can now drop a lot of cruft and blocking transitions. The performance numbers vary depending on the workload, because generally speaking there doesn't tend to be a lot of contention on the btree. However, on my test system which is an 80 core single socket system with 256GiB of RAM and a 2TiB NVMe drive I get the following results (with all debug options off): dbench 200 baseline Throughput 216.056 MB/sec 200 clients 200 procs max_latency=1471.197 ms dbench 200 with patch Throughput 737.188 MB/sec 200 clients 200 procs max_latency=714.346 ms Previously we also used fs_mark to test this sort of contention, and those results are far less impressive, mostly because there's not enough tasks to really stress the locking fs_mark -d /d[0-15] -S 0 -L 20 -n 100000 -s 0 -t 16 baseline Average Files/sec: 160166.7 p50 Files/sec: 165832 p90 Files/sec: 123886 p99 Files/sec: 123495 real 3m26.527s user 2m19.223s sys 48m21.856s patched Average Files/sec: 164135.7 p50 Files/sec: 171095 p90 Files/sec: 122889 p99 Files/sec: 113819 real 3m29.660s user 2m19.990s sys 44m12.259s Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba

btrfs: introduce BTRFS_NESTING_NEW_ROOT for adding new roots

2020-10-07T10:12:17+00:00

The way we add new roots is confusing from a locking perspective for lockdep. We generally have the rule that we lock things in order from highest level to lowest, but in the case of adding a new level to the tree we actually allocate a new block for the root, which makes the locking go in reverse. A similar issue exists for snapshotting, we cow the original root for the root of a new tree, however they're at the same level. Address this by using BTRFS_NESTING_NEW_ROOT for these operations. Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba

btrfs: introduce BTRFS_NESTING_SPLIT for split blocks

2020-10-07T10:12:16+00:00

If we are splitting a leaf/node, we could do something like the following lock(leaf) BTRFS_NESTING_NORMAL lock(left) BTRFS_NESTING_LEFT + BTRFS_NESTING_COW push from leaf -> left reset path to point to left split left allocate new block, lock block BTRFS_NESTING_SPLIT at the new block point we need to have a different nesting level, because we have already used either BTRFS_NESTING_LEFT or BTRFS_NESTING_RIGHT when pushing items from the original leaf into the adjacent leaves. Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba

btrfs: introduce BTRFS_NESTING_LEFT/RIGHT_COW

2020-10-07T10:12:16+00:00

For similar reasons as BTRFS_NESTING_COW, we need BTRFS_NESTING_LEFT/RIGHT_COW. The pattern is this lock leaf -> BTRFS_NESTING_NORMAL cow leaf -> BTRFS_NESTING_COW split leaf lock left -> BTRFS_NESTING_LEFT cow left -> BTRFS_NESTING_LEFT_COW We need this in order to indicate to lockdep that these locks are discrete and are being taken in a safe order. Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba

btrfs: introduce BTRFS_NESTING_LEFT/BTRFS_NESTING_RIGHT

2020-10-07T10:12:16+00:00

Our lockdep maps are based on rootid+level, however in some cases we will lock adjacent blocks on the same level, namely in searching forward or in split/balance. Because of this lockdep will complain, so we need a separate subclass to indicate to lockdep that these are different locks. lock leaf -> BTRFS_NESTING_NORMAL cow leaf -> BTRFS_NESTING_COW split leaf lock left -> BTRFS_NESTING_LEFT lock right -> BTRFS_NESTING_RIGHT The above graph illustrates the need for this new nesting subclass. Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba

btrfs: introduce BTRFS_NESTING_COW for cow'ing blocks

2020-10-07T10:12:16+00:00

When we COW a block we are holding a lock on the original block, and then we lock the new COW block. Because our lockdep maps are based on root + level, this will make lockdep complain. We need a way to indicate a subclass for locking the COW'ed block, so plumb through our btrfs_lock_nesting from btrfs_cow_block down to the btrfs_init_buffer, and then introduce BTRFS_NESTING_COW to be used for cow'ing blocks. The reason I've added all this extra infrastructure is because there will be need of different nesting classes in follow up patches. Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba

btrfs: add nesting tags to the locking helpers

2020-10-07T10:12:16+00:00

We will need these when we switch to an rwsem, so plumb in the infrastructure here to use later on. I violate the 80 character limit some here because it'll be cleaned up later. Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba