In commit b150a4f10d878 ("Btrfs: use a percpu to keep track of
possibly
pinned bytes") we use total_bytes_pinned to track how many bytes we
are
going to free in this transaction. When we are close to ENOSPC, we
check it
and know if we can make the allocation by commit the current
transaction.
For every data/metadata extent we are going to free, we add
total_bytes_pinned in btrfs_free_extent() and btrfs_free_tree_block(),
and
release it in unpin_extent_range() when we finish the transaction. So
this
is a variable we frequently update but rarely read - just the suitable
use of percpu_counter. But in previous commit we update
total_bytes_pinned
by default 32 batch size, making every update essentially a spin lock
protected update. Since every spin lock/unlock operation involves
syncing
a globally used variable and some kind of barrier in a SMP system,
this is
more expensive than using total_bytes_pinned as a simple atomic64_t.
So
fix this by using a customized batch size. Since we only read
total_bytes_pinned when we are close to ENOSPC and fail to alloc new
chunk,
we can use a really large batch size and have nearly no penalty in
most
cases.
[Test]
We test the patch on a 4-cores x86 machine:
1. falloate a 16GiB size test file.
2. take snapshot (so all following writes will be cow write).
3. run a 180 sec, 4 jobs, 4K random write fio on test file.
We also add a temporary lockdep class on percpu_counter's spin lock
used
by total_bytes_pinned to track lock_stat.
[Results]
unpatched:
lock_stat version 0.4
-----------------------------------------------------------------------
class name con-bounces contentions
waittime-min waittime-max waittime-total waittime-avg
acq-bounces
acquisitions holdtime-min holdtime-max holdtime-total
holdtime-avg
total_bytes_pinned_percpu: 82 82
0.21 0.61 29.46 0.36
298340
635973 0.09 11.01 173476.25
0.27
patched:
lock_stat version 0.4
-----------------------------------------------------------------------
class name con-bounces contentions
waittime-min waittime-max waittime-total waittime-avg
acq-bounces
acquisitions holdtime-min holdtime-max holdtime-total
holdtime-avg
total_bytes_pinned_percpu: 1 1
0.62 0.62 0.62 0.62
13601
31542 0.14 9.61 11016.90
0.35
[Analysis]
Since the spin lock only protect a single in-memory variable, the
contentions (number of lock acquisitions that had to wait) in both
unpatched and patched version are low. But when we see acquisitions
and
acq-bounces, we get much lower counts in patched version. Here the
most
important metric is acq-bounces. It means how many times the lock get
transferred between different cpus, so the patch can really recude
cacheline bouncing of spin lock (also the global counter of
percpu_counter)
in a SMP system.
Fixes: b150a4f10d878 ("Btrfs: use a percpu to keep track of possibly
pinned bytes")
Signed-off-by: Ethan Lien <ethanl...@synology.com>
---
V2:
Rewrite commit comments.
Add lock_stat test.
Pull dirty_metadata_bytes out to a separate patch.
fs/btrfs/ctree.h | 1 +
fs/btrfs/extent-tree.c | 46
++++++++++++++++++++++++++++--------------
2 files changed, 32 insertions(+), 15 deletions(-)
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 118346aceea9..df682a521635 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -422,6 +422,7 @@ struct btrfs_space_info {
* time the transaction commits.
*/
struct percpu_counter total_bytes_pinned;
+ s32 total_bytes_pinned_batch;
