From: Dave Chinner <dchinner@xxxxxxxxxx>
Now that the per-sb shrinker is responsible for shrinking 2 or more
caches, increase the batch size to keep econmies of scale for
shrinking each cache. Increase the shrinker batch size to 1024
To allow for a large increase in batch size, add a conditional
reschedule to prune_icache_sb() so that we don't hold the LRU spin
lock for too long. This mirrors the behaviour of the
__shrink_dcache_sb(), and allows us to increase the batch size
without needing to worry about problems caused by long lock hold
To ensure that filesystems using the per-sb shrinker callouts don't
cause problems, document that the object freeing method must
reschedule appropriately inside loops.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
Documentation/filesystems/vfs.txt | 5 +++++
fs/super.c | 1 +
2 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/Documentation/filesystems/vfs.txt
index dc732d2..2e26973 100644
@@ -317,6 +317,11 @@ or bottom half).
the VM is trying to reclaim under GFP_NOFS conditions, hence this
method does not need to handle that situation itself.
+ Implementations must include conditional reschedule calls inside any
+ scanning loop that is done. This allows the VFS to determine
+ appropriate scan batch sizes without having to worry about whether
+ implementations will cause holdoff problems due ot large batch sizes.
Whoever sets up the inode is responsible for filling in the "i_op" field. This
is a pointer to a "struct inode_operations" which describes the methods that
can be performed on individual inodes.
diff --git a/fs/super.c b/fs/super.c
index b55f968..323a63e 100644
@@ -184,6 +184,7 @@ static struct super_block *alloc_super(struct
s->s_shrink.seeks = DEFAULT_SEEKS;
s->s_shrink.shrink = prune_super;
+ s->s_shrink.batch = 1024;