From: Kari Argillander <kari.argillander@gmail.com>
To: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
Cc: ntfs3@lists.linux.dev, linux-kernel@vger.kernel.org,
linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH 1/5] fs/ntfs3: Move ni_lock_dir and ni_unlock into ntfs_create_inode
Date: Wed, 22 Sep 2021 21:12:44 +0300 [thread overview]
Message-ID: <20210922181244.tjxij5gi6xft4wwh@kari-VirtualBox> (raw)
In-Reply-To: <a269be2f-4ab6-b1c6-790c-9d3052bf22cc@paragon-software.com>
On Wed, Sep 22, 2021 at 07:17:13PM +0300, Konstantin Komarov wrote:
> Now ntfs3 locks mutex for smaller time.
Really like this change. It was my todo list also. Thanks. Still some
comments below.
>
> Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
> ---
> fs/ntfs3/inode.c | 17 ++++++++++++++---
> fs/ntfs3/namei.c | 20 --------------------
> 2 files changed, 14 insertions(+), 23 deletions(-)
>
> diff --git a/fs/ntfs3/inode.c b/fs/ntfs3/inode.c
> index d583b71bec50..6fc99eebd1c1 100644
> --- a/fs/ntfs3/inode.c
> +++ b/fs/ntfs3/inode.c
> @@ -1198,9 +1198,13 @@ struct inode *ntfs_create_inode(struct user_namespace *mnt_userns,
> struct REPARSE_DATA_BUFFER *rp = NULL;
> bool rp_inserted = false;
>
> + ni_lock_dir(dir_ni);
> +
> dir_root = indx_get_root(&dir_ni->dir, dir_ni, NULL, NULL);
> - if (!dir_root)
> - return ERR_PTR(-EINVAL);
> + if (!dir_root) {
> + err = -EINVAL;
> + goto out1;
> + }
>
> if (S_ISDIR(mode)) {
> /* Use parent's directory attributes. */
> @@ -1549,6 +1553,9 @@ struct inode *ntfs_create_inode(struct user_namespace *mnt_userns,
> if (err)
> goto out6;
>
> + /* Unlock parent directory before ntfs_init_acl. */
> + ni_unlock(dir_ni);
There is err path which can go to err6 (line 1585). Then we get double
unlock situation.
> +
> inode->i_generation = le16_to_cpu(rec->seq);
>
> dir->i_mtime = dir->i_ctime = inode->i_atime;
> @@ -1605,8 +1612,10 @@ struct inode *ntfs_create_inode(struct user_namespace *mnt_userns,
> out7:
>
> /* Undo 'indx_insert_entry'. */
> + ni_lock_dir(dir_ni);
> indx_delete_entry(&dir_ni->dir, dir_ni, new_de + 1,
> le16_to_cpu(new_de->key_size), sbi);
> + /* ni_unlock(dir_ni); will be called later. */
> out6:
> if (rp_inserted)
> ntfs_remove_reparse(sbi, IO_REPARSE_TAG_SYMLINK, &new_de->ref);
> @@ -1630,8 +1639,10 @@ struct inode *ntfs_create_inode(struct user_namespace *mnt_userns,
> kfree(rp);
>
> out1:
> - if (err)
> + if (err) {
> + ni_unlock(dir_ni);
This will be double unlock if we exit with err path out6.
Argillander
> return ERR_PTR(err);
> + }
>
> unlock_new_inode(inode);
>
> diff --git a/fs/ntfs3/namei.c b/fs/ntfs3/namei.c
> index 1c475da4e19d..bc741213ad84 100644
> --- a/fs/ntfs3/namei.c
> +++ b/fs/ntfs3/namei.c
> @@ -95,16 +95,11 @@ static struct dentry *ntfs_lookup(struct inode *dir, struct dentry *dentry,
> static int ntfs_create(struct user_namespace *mnt_userns, struct inode *dir,
> struct dentry *dentry, umode_t mode, bool excl)
> {
> - struct ntfs_inode *ni = ntfs_i(dir);
> struct inode *inode;
>
> - ni_lock_dir(ni);
> -
> inode = ntfs_create_inode(mnt_userns, dir, dentry, NULL, S_IFREG | mode,
> 0, NULL, 0, NULL);
>
> - ni_unlock(ni);
> -
> return IS_ERR(inode) ? PTR_ERR(inode) : 0;
> }
>
> @@ -116,16 +111,11 @@ static int ntfs_create(struct user_namespace *mnt_userns, struct inode *dir,
> static int ntfs_mknod(struct user_namespace *mnt_userns, struct inode *dir,
> struct dentry *dentry, umode_t mode, dev_t rdev)
> {
> - struct ntfs_inode *ni = ntfs_i(dir);
> struct inode *inode;
>
> - ni_lock_dir(ni);
> -
> inode = ntfs_create_inode(mnt_userns, dir, dentry, NULL, mode, rdev,
> NULL, 0, NULL);
>
> - ni_unlock(ni);
> -
> return IS_ERR(inode) ? PTR_ERR(inode) : 0;
> }
>
> @@ -196,15 +186,10 @@ static int ntfs_symlink(struct user_namespace *mnt_userns, struct inode *dir,
> {
> u32 size = strlen(symname);
> struct inode *inode;
> - struct ntfs_inode *ni = ntfs_i(dir);
> -
> - ni_lock_dir(ni);
>
> inode = ntfs_create_inode(mnt_userns, dir, dentry, NULL, S_IFLNK | 0777,
> 0, symname, size, NULL);
>
> - ni_unlock(ni);
> -
> return IS_ERR(inode) ? PTR_ERR(inode) : 0;
> }
>
> @@ -215,15 +200,10 @@ static int ntfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir,
> struct dentry *dentry, umode_t mode)
> {
> struct inode *inode;
> - struct ntfs_inode *ni = ntfs_i(dir);
> -
> - ni_lock_dir(ni);
>
> inode = ntfs_create_inode(mnt_userns, dir, dentry, NULL, S_IFDIR | mode,
> 0, NULL, 0, NULL);
>
> - ni_unlock(ni);
> -
> return IS_ERR(inode) ? PTR_ERR(inode) : 0;
> }
>
> --
> 2.33.0
>
>
next prev parent reply other threads:[~2021-09-22 18:12 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-22 16:15 [PATCH 0/5] Refactor locking in inode_operations Konstantin Komarov
2021-09-22 16:17 ` [PATCH 1/5] fs/ntfs3: Move ni_lock_dir and ni_unlock into ntfs_create_inode Konstantin Komarov
2021-09-22 18:12 ` Kari Argillander [this message]
2021-09-22 16:18 ` [PATCH 2/5] fs/ntfs3: Refactor ntfs_get_acl_ex for better readability Konstantin Komarov
2021-09-22 17:47 ` Kari Argillander
2021-09-22 16:19 ` [PATCH 3/5] fs/ntfs3: Pass flags to ntfs_set_ea in ntfs_set_acl_ex Konstantin Komarov
2021-09-22 17:59 ` Kari Argillander
2021-09-22 16:20 ` [PATCH 4/5] fs/ntfs3: Change posix_acl_equiv_mode to posix_acl_update_mode Konstantin Komarov
2021-09-22 18:23 ` Kari Argillander
2021-09-22 16:20 ` [PATCH 5/5] fs/ntfs3: Refactoring lock in ntfs_init_acl Konstantin Komarov
2021-09-22 18:41 ` Kari Argillander
2021-09-22 18:51 ` [PATCH 0/5] Refactor locking in inode_operations Kari Argillander
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=20210922181244.tjxij5gi6xft4wwh@kari-VirtualBox \
--to=kari.argillander@gmail.com \
--cc=almaz.alexandrovich@paragon-software.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=ntfs3@lists.linux.dev \
/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
all inboxes | Powered by JetHome®