diff mbox series

[v3,12/12] btrfs: enable nowait async buffered writes

Message ID 20220912192752.3785061-13-shr@fb.com (mailing list archive)
State New
Headers show
Series io-uring/btrfs: support async buffered writes | expand

Commit Message

Stefan Roesch Sept. 12, 2022, 7:27 p.m. UTC
Enable nowait async buffered writes in btrfs_do_write_iter() and
btrfs_file_open().

In this version encoded buffered writes have the optimization not
enabled. Encoded writes are enabled by using an ioctl. io-uring
currently do not support ioctl's. This might be enabled in the future.

Performance results:
  For fio the following results have been obtained with a queue depth of
  1 and 4k block size (runtime 600 secs):

                 sequential writes:
                 without patch           with patch      libaio     psync
  iops:              55k                    134k          117K       148K
  bw:               221MB/s                 538MB/s       469MB/s    592MB/s
  clat:           15286ns                    82ns         994ns     6340ns

For an io depth of 1, the new patch improves throughput by over two times
(compared to the exiting behavior, where buffered writes are processed by an
io-worker process) and also the latency is considerably reduced. To achieve the
same or better performance with the exisiting code an io depth of 4 is required.
Increasing the iodepth further does not lead to improvements.

The tests have been run like this:
./fio --name=seq-writers --ioengine=psync --iodepth=1 --rw=write \
  --bs=4k --direct=0 --size=100000m --time_based --runtime=600   \
  --numjobs=1 --filename=...
./fio --name=seq-writers --ioengine=io_uring --iodepth=1 --rw=write \
  --bs=4k --direct=0 --size=100000m --time_based --runtime=600   \
  --numjobs=1 --filename=...
./fio --name=seq-writers --ioengine=libaio --iodepth=1 --rw=write \
  --bs=4k --direct=0 --size=100000m --time_based --runtime=600   \
  --numjobs=1 --filename=...

Testing:
  This patch has been tested with xfstests, fsx, fio. xfstests shows no new
  diffs compared to running without the patch series.

Signed-off-by: Stefan Roesch <shr@fb.com>
---
 fs/btrfs/file.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
diff mbox series

Patch

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 4dc6484ff229..16052903fa82 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -2107,13 +2107,13 @@  ssize_t btrfs_do_write_iter(struct kiocb *iocb, struct iov_iter *from,
 	if (BTRFS_FS_ERROR(inode->root->fs_info))
 		return -EROFS;
 
-	if ((iocb->ki_flags & IOCB_NOWAIT) && !(iocb->ki_flags & IOCB_DIRECT))
-		return -EOPNOTSUPP;
-
 	if (sync)
 		atomic_inc(&inode->sync_writers);
 
 	if (encoded) {
+		if (iocb->ki_flags & IOCB_NOWAIT)
+			return -EOPNOTSUPP;
+
 		num_written = btrfs_encoded_write(iocb, from, encoded);
 		num_sync = encoded->len;
 	} else if (iocb->ki_flags & IOCB_DIRECT) {
@@ -3755,7 +3755,7 @@  static int btrfs_file_open(struct inode *inode, struct file *filp)
 {
 	int ret;
 
-	filp->f_mode |= FMODE_NOWAIT | FMODE_BUF_RASYNC;
+	filp->f_mode |= FMODE_NOWAIT | FMODE_BUF_RASYNC | FMODE_BUF_WASYNC;
 
 	ret = fsverity_file_open(inode, filp);
 	if (ret)