[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20250317133611.457-1-yohan.joung@sk.com>
Date: Mon, 17 Mar 2025 22:36:11 +0900
From: Yohan Joung <jyh429@...il.com>
To: jaegeuk@...nel.org,
chao@...nel.org,
daeho43@...il.com
Cc: linux-f2fs-devel@...ts.sourceforge.net,
linux-kernel@...r.kernel.org,
Yohan Joung <yohan.joung@...com>
Subject: [PATCH v2] f2fs: optimize f2fs DIO overwrites
this is unnecessary when we know we are overwriting already allocated
blocks and the overhead of starting a transaction can be significant
especially for multithreaded workloads doing small writes.
Signed-off-by: Yohan Joung <yohan.joung@...com>
---
fs/f2fs/data.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index d872f785a996..d99ec9370b1d 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -4181,7 +4181,13 @@ static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
map.m_next_pgofs = &next_pgofs;
map.m_seg_type = f2fs_rw_hint_to_seg_type(F2FS_I_SB(inode),
inode->i_write_hint);
- if (flags & IOMAP_WRITE)
+
+ /*
+ * If the blocks being overwritten are already allocated,
+ * f2fs_map_lock and f2fs_balance_fs are not necessary.
+ */
+ if ((flags & IOMAP_WRITE) &&
+ !f2fs_overwrite_io(inode, offset, length))
map.m_may_create = true;
err = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_DIO);
--
2.25.1
Powered by blists - more mailing lists