Re: [PATCH 4/4] fs/ntfs3: Optimize locking in ntfs_save_wsl_perm

[Kari Argillander <kari.argillander@gmail.com>]

On Tue, Oct 26, 2021 at 07:42:15PM +0300, Konstantin Komarov wrote:
> Right now in ntfs_save_wsl_perm we lock/unlock 4 times.
> This commit fixes this situation.
> We add "locked" argument to ntfs_set_ea.
> 
> Suggested-by: Kari Argillander <kari.argillander@gmail.com>
> Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>

Add tag if you fix Joes nit.

Reviewed-by: Kari Argillander <kari.argillander@gmail.com>

> ---
>  fs/ntfs3/xattr.c | 24 ++++++++++++++----------
>  1 file changed, 14 insertions(+), 10 deletions(-)
> 
> diff --git a/fs/ntfs3/xattr.c b/fs/ntfs3/xattr.c
> index 157b70aecb4f..6d8b1cd7681d 100644
> --- a/fs/ntfs3/xattr.c
> +++ b/fs/ntfs3/xattr.c
> @@ -259,7 +259,7 @@ static int ntfs_get_ea(struct inode *inode, const char *name, size_t name_len,
>  
>  static noinline int ntfs_set_ea(struct inode *inode, const char *name,
>  				size_t name_len, const void *value,
> -				size_t val_size, int flags)
> +				size_t val_size, int flags, bool locked)
>  {
>  	struct ntfs_inode *ni = ntfs_i(inode);
>  	struct ntfs_sb_info *sbi = ni->mi.sbi;
> @@ -278,7 +278,8 @@ static noinline int ntfs_set_ea(struct inode *inode, const char *name,
>  	u64 new_sz;
>  	void *p;
>  
> -	ni_lock(ni);
> +	if (!locked)
> +		ni_lock(ni);
>  
>  	run_init(&ea_run);
>  
> @@ -467,7 +468,8 @@ static noinline int ntfs_set_ea(struct inode *inode, const char *name,
>  	mark_inode_dirty(&ni->vfs_inode);
>  
>  out:
> -	ni_unlock(ni);
> +	if (!locked)
> +		ni_unlock(ni);
>  
>  	run_close(&ea_run);
>  	kfree(ea_all);
> @@ -595,7 +597,7 @@ static noinline int ntfs_set_acl_ex(struct user_namespace *mnt_userns,
>  		flags = 0;
>  	}
>  
> -	err = ntfs_set_ea(inode, name, name_len, value, size, flags);
> +	err = ntfs_set_ea(inode, name, name_len, value, size, flags, 0);
>  	if (err == -ENODATA && !size)
>  		err = 0; /* Removing non existed xattr. */
>  	if (!err)
> @@ -989,7 +991,7 @@ static noinline int ntfs_setxattr(const struct xattr_handler *handler,
>  	}
>  #endif
>  	/* Deal with NTFS extended attribute. */
> -	err = ntfs_set_ea(inode, name, name_len, value, size, flags);
> +	err = ntfs_set_ea(inode, name, name_len, value, size, flags, 0);

Did you miss Joes comment or is there another reason why there is still
is (true|0) arguments?

https://lore.kernel.org/ntfs3/adcb168fc78f62583f8d925bcadbbcda9ba7da20.camel@perches.com/

  argillander

>  
>  out:
>  	inode->i_ctime = current_time(inode);
> @@ -1007,35 +1009,37 @@ int ntfs_save_wsl_perm(struct inode *inode)
>  {
>  	int err;
>  	__le32 value;
> +	struct ntfs_inode *ni = ntfs_i(inode);
>  
> -	/* TODO: refactor this, so we don't lock 4 times in ntfs_set_ea */
> +	ni_lock(ni);
>  	value = cpu_to_le32(i_uid_read(inode));
>  	err = ntfs_set_ea(inode, "$LXUID", sizeof("$LXUID") - 1, &value,
> -			  sizeof(value), 0);
> +			  sizeof(value), 0, true); /* true == already locked. */
>  	if (err)
>  		goto out;
>  
>  	value = cpu_to_le32(i_gid_read(inode));
>  	err = ntfs_set_ea(inode, "$LXGID", sizeof("$LXGID") - 1, &value,
> -			  sizeof(value), 0);
> +			  sizeof(value), 0, true);
>  	if (err)
>  		goto out;
>  
>  	value = cpu_to_le32(inode->i_mode);
>  	err = ntfs_set_ea(inode, "$LXMOD", sizeof("$LXMOD") - 1, &value,
> -			  sizeof(value), 0);
> +			  sizeof(value), 0, true);
>  	if (err)
>  		goto out;
>  
>  	if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) {
>  		value = cpu_to_le32(inode->i_rdev);
>  		err = ntfs_set_ea(inode, "$LXDEV", sizeof("$LXDEV") - 1, &value,
> -				  sizeof(value), 0);
> +				  sizeof(value), 0, true);
>  		if (err)
>  			goto out;
>  	}
>  
>  out:
> +	ni_unlock(ni);
>  	/* In case of error should we delete all WSL xattr? */
>  	return err;
>  }
> -- 
> 2.33.0
> 
> 
>