Skip to contentSkip to conversation

nilfs2: remove redundant super block commit

PATCH Applied Jiro SEKIBA 2 participants 2 messages Jul 10, 2009 1 message not in the archive NILFS2 filesystem

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 nilfs2: remove redundant super block commit

Hi,

This is a patch to remove redundant super block commit.

nilfs_write_super will call nilfs_commit_super to store super block into block device. However, nilfs_put_super will call nilfs_commit_super right after calling nilfs_write_super.
So calling nilfs_write_super in nilfs_put_super would be redundant.

thanks,

regards,

Signature · 7 lines
Jiro SEKIBA <[email protected]>

 Signed-off-by: Jiro SEKIBA <[email protected]>

---
 fs/nilfs2/super.c |    3 ---
 1 files changed, 0 insertions(+), 3 deletions(-)
fs/nilfs2/super.c +0 -3
index b1d992b..52c089f 100644
@@ -313,9 +313,6 @@ static void nilfs_put_super(struct super_block *sb)
 
 	lock_kernel();
 
-	if (sb->s_dirt)
-		nilfs_write_super(sb);
-
 	nilfs_detach_segment_constructor(sbi);
 
 	if (!(sb->s_flags & MS_RDONLY)) {
Signature · 1 line
1.5.6.5

A message that is not in the archive (sent off-list or not yet received): 87ljmxruns.wl%[email protected]

#2 lore raw

Hi,

Quoted 4 lines · Jiro SEKIBA

Hi,

This is a patch to remove redundant super block commit.

Quoted 12 lines

nilfs_write_super will call nilfs_commit_super to store super block into block device. However, nilfs_put_super will call nilfs_commit_super right after calling nilfs_write_super.
So calling nilfs_write_super in nilfs_put_super would be redundant.

thanks,

regards,

Signature · 3 lines
Jiro SEKIBA <[email protected]>

 Signed-off-by: Jiro SEKIBA <[email protected]>

Applied. Thanks!

Due to the recent changes on vfs interface, some functions in fs/nilfs2/super.c has room for cleanup like this.

Further, there is a discussion about eliminating the write_super() callback in LKML. The function was periodically called from the vfs layer to induce a writeback of superblock. It is now deemed not to be required if a file system handles the timing. I feel the writeback control of the superblock should be revised overall for nilfs.

I would appreciate it if you could help me to do the series of cleanup jobs.

Thanks,
Ryusuke Konishi

Quoted 22 lines
---
 fs/nilfs2/super.c |    3 ---
 1 files changed, 0 insertions(+), 3 deletions(-)
fs/nilfs2/super.c +0 -3
index b1d992b..52c089f 100644
@@ -313,9 +313,6 @@ static void nilfs_put_super(struct super_block *sb)
 
 	lock_kernel();
 
-	if (sb->s_dirt)
-		nilfs_write_super(sb);
-
 	nilfs_detach_segment_constructor(sbi);
 
 	if (!(sb->s_flags & MS_RDONLY)) {
Signature · 1 line
1.5.6.5