linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Josef Bacik <jbacik@fb.com>
To: <linux-btrfs@vger.kernel.org>
Subject: [PATCH 14/14] Btrfs: don't do nocow check unless we have to
Date: Fri, 25 Mar 2016 13:26:00 -0400	[thread overview]
Message-ID: <1458926760-17563-15-git-send-email-jbacik@fb.com> (raw)
In-Reply-To: <1458926760-17563-1-git-send-email-jbacik@fb.com>

Before we write into prealloc/nocow space we have to make sure that there are no
references to the extents we are writing into, which means checking the extent
tree and csum tree in the case of nocow.  So we don't want to do the nocow dance
unless we can't reserve data space, since it's a serious drag on performance.
With the following sequence

fallocate -l10737418240 /mnt/btrfs-test/file
cp --reflink /mnt/btrfs-test/file /mnt/btrfs-test/link
fio --name=randwrite --rw=randwrite --bs=4k --filename=/mnt/btrfs-test/file \
	--end_fsync=1

we get the worst case scenario where we have to fall back on to doing the check
anyway.

Without this patch
lat (usec): min=5, max=111598, avg=27.65, stdev=124.51
write: io=10240MB, bw=126876KB/s, iops=31718, runt= 82646msec

With this patch
lat (usec): min=3, max=91210, avg=14.09, stdev=110.62
write: io=10240MB, bw=212753KB/s, iops=53188, runt= 49286msec

We get twice the throughput, half of the runtime, and half of the average
latency.  Thanks,

Signed-off-by: Josef Bacik <jbacik@fb.com>
---
 fs/btrfs/file.c | 44 ++++++++++++++++++++++----------------------
 1 file changed, 22 insertions(+), 22 deletions(-)

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 0ce4bb3..7c80208 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1534,30 +1534,30 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
 		reserve_bytes = round_up(write_bytes + sector_offset,
 				root->sectorsize);
 
-		if ((BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW |
-					      BTRFS_INODE_PREALLOC)) &&
-		    check_can_nocow(inode, pos, &write_bytes) > 0) {
-			/*
-			 * For nodata cow case, no need to reserve
-			 * data space.
-			 */
-			only_release_metadata = true;
-			/*
-			 * our prealloc extent may be smaller than
-			 * write_bytes, so scale down.
-			 */
-			num_pages = DIV_ROUND_UP(write_bytes + offset,
-						 PAGE_CACHE_SIZE);
-			reserve_bytes = round_up(write_bytes + sector_offset,
-					root->sectorsize);
-			goto reserve_metadata;
-		}
-
 		ret = btrfs_check_data_free_space(inode, pos, write_bytes);
-		if (ret < 0)
-			break;
+		if (ret < 0) {
+			if ((BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW |
+						      BTRFS_INODE_PREALLOC)) &&
+			    check_can_nocow(inode, pos, &write_bytes) > 0) {
+				/*
+				 * For nodata cow case, no need to reserve
+				 * data space.
+				 */
+				only_release_metadata = true;
+				/*
+				 * our prealloc extent may be smaller than
+				 * write_bytes, so scale down.
+				 */
+				num_pages = DIV_ROUND_UP(write_bytes + offset,
+							 PAGE_CACHE_SIZE);
+				reserve_bytes = round_up(write_bytes +
+							 sector_offset,
+							 root->sectorsize);
+			} else {
+				break;
+			}
+		}
 
-reserve_metadata:
 		ret = btrfs_delalloc_reserve_metadata(inode, reserve_bytes);
 		if (ret) {
 			if (!only_release_metadata)
-- 
2.5.0


  parent reply	other threads:[~2016-03-25 17:26 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-03-25 17:25 [PATCH 00/14] Enospc rework Josef Bacik
2016-03-25 17:25 ` [PATCH 01/14] Btrfs: add bytes_readonly to the spaceinfo at once Josef Bacik
2016-03-25 17:25 ` [PATCH 02/14] Btrfs: fix callers of btrfs_block_rsv_migrate Josef Bacik
2016-03-25 17:25 ` [PATCH 03/14] Btrfs: always reserve metadata for delalloc extents Josef Bacik
2016-03-25 18:04   ` Liu Bo
2016-03-25 17:25 ` [PATCH 04/14] Btrfs: change delayed reservation fallback behavior Josef Bacik
2016-03-25 17:25 ` [PATCH 05/14] Btrfs: warn_on for unaccounted spaces Josef Bacik
2016-06-27  4:47   ` Qu Wenruo
2016-06-27 13:03     ` Chris Mason
2016-06-28  0:16       ` Qu Wenruo
2016-03-25 17:25 ` [PATCH 06/14] Btrfs: add tracepoint for adding block groups Josef Bacik
2016-03-25 17:25 ` [PATCH 07/14] Btrfs: introduce ticketed enospc infrastructure Josef Bacik
2016-05-09 21:29   ` Liu Bo
2016-05-17 17:30   ` [PATCH V2] " Josef Bacik
2016-05-18 11:24     ` Austin S. Hemmelgarn
2016-05-19 12:47       ` Austin S. Hemmelgarn
2016-05-18 22:46     ` David Sterba
2016-03-25 17:25 ` [PATCH 08/14] Btrfs: trace pinned extents Josef Bacik
2016-03-25 17:25 ` [PATCH 09/14] Btrfs: fix delalloc reservation amount tracepoint Josef Bacik
2016-03-25 17:25 ` [PATCH 10/14] Btrfs: add tracepoints for flush events Josef Bacik
2016-03-25 17:25 ` [PATCH 11/14] Btrfs: add fsid to some tracepoints Josef Bacik
2016-03-25 17:25 ` [PATCH 12/14] Btrfs: fix release reserved extents trace points Josef Bacik
2016-05-09 21:33   ` Liu Bo
2016-03-25 17:25 ` [PATCH 13/14] Btrfs: don't bother kicking async if there's nothing to reclaim Josef Bacik
2016-03-25 17:26 ` Josef Bacik [this message]
2016-03-25 17:50   ` [PATCH 14/14] Btrfs: don't do nocow check unless we have to Liu Bo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1458926760-17563-15-git-send-email-jbacik@fb.com \
    --to=jbacik@fb.com \
    --cc=linux-btrfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).