Btrfs: avoid allocation clusters that are too spread out
In SSD mode for data, and all the time for metadata the allocator
will try to find a cluster of nearby blocks for allocations. This
commit adds extra checks to make sure that each free block in the
cluster is close to the last one.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 28016558..ac23476 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -645,7 +645,8 @@
* we haven't filled the empty size and the window is
* very large. reset and try again
*/
- if (next->offset - window_start > (bytes + empty_size) * 2) {
+ if (next->offset - (last->offset + last->bytes) > 128 * 1024 ||
+ next->offset - window_start > (bytes + empty_size) * 2) {
entry = next;
window_start = entry->offset;
window_free = entry->bytes;