From: Amir Goldstein <amir73il@gmail.com>
To: Miklos Szeredi <miklos@szeredi.hu>
Cc: linux-unionfs@vger.kernel.org
Subject: [PATCH 01/17] vfs: add helper wait_on_inode_inuse()
Date: Fri, 2 Jun 2017 17:04:28 +0300 [thread overview]
Message-ID: <1496412284-4113-2-git-send-email-amir73il@gmail.com> (raw)
In-Reply-To: <1496412284-4113-1-git-send-email-amir73il@gmail.com>
Adds a helper to wait until an inode's INUSE bit is cleared.
This is going to be used by overlayfs to sychronize concurrent copy up
of lower hardlinks.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
---
fs/inode.c | 24 ++++++++++++++++++++++++
include/linux/fs.h | 1 +
2 files changed, 25 insertions(+)
diff --git a/fs/inode.c b/fs/inode.c
index 216efeecdbdc..546cd503148a 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -2167,6 +2167,30 @@ void inode_inuse_unlock(struct inode *inode)
WARN_ON(inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE));
WARN_ON(!(inode->i_state & I_INUSE));
inode->i_state &= ~I_INUSE;
+ smp_mb();
+ wake_up_bit(&inode->i_state, __I_INUSE);
spin_unlock(&inode->i_lock);
}
EXPORT_SYMBOL(inode_inuse_unlock);
+
+/**
+ * wait_on_inode_inuse - wait for release of exclusive 'inuse' lock
+ * @inode: inode inuse to wait on
+ *
+ * Can be used in combination with parent i_mutex, to protect access to a
+ * newly created inode, until that inode has been properly initialized by
+ * the user that grabbed the 'inuse' exclusive lock after creating the inode.
+ *
+ * Caller must hold a reference to inode to prevent waiting on an inode that
+ * is not 'inuse' and is already being freed.
+ *
+ * Return 0 if the 'inuse' bit is clear or has been cleared while waiting.
+ */
+int wait_on_inode_inuse(struct inode *inode, unsigned mode)
+{
+ if (WARN_ON(!atomic_read(&inode->i_count)))
+ return -EINVAL;
+ might_sleep();
+ return wait_on_bit(&inode->i_state, __I_INUSE, mode);
+}
+EXPORT_SYMBOL(wait_on_inode_inuse);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 2e29ff868e90..e064612b45ef 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -3268,6 +3268,7 @@ extern bool path_noexec(const struct path *path);
extern void inode_nohighmem(struct inode *inode);
extern bool inode_inuse_trylock(struct inode *inode);
extern void inode_inuse_unlock(struct inode *inode);
+extern int wait_on_inode_inuse(struct inode *inode, unsigned mode);
static inline bool inode_inuse(struct inode *inode)
{
--
2.7.4
next prev parent reply other threads:[~2017-06-02 14:04 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-02 14:04 [PATCH 00/17] Avoid breaking lower hardlinks on copy up Amir Goldstein
2017-06-02 14:04 ` Amir Goldstein [this message]
2017-06-02 14:04 ` [PATCH 02/17] ovl: generalize ovl_create_workdir() Amir Goldstein
2017-06-02 14:04 ` [PATCH 03/17] ovl: introduce the inodes index dir feature Amir Goldstein
2017-06-02 14:04 ` [PATCH 04/17] ovl: verify index dir matches upper dir Amir Goldstein
2017-06-02 14:04 ` [PATCH 05/17] ovl: create helper ovl_lookup_index() Amir Goldstein
2017-06-02 14:04 ` [PATCH 06/17] ovl: move inode helpers to inode.c Amir Goldstein
2017-06-02 14:04 ` [PATCH 07/17] ovl: create helpers for initializing hashed inode Amir Goldstein
2017-06-02 14:04 ` [PATCH 08/17] ovl: use ovl_inode_init() for initializing new inode Amir Goldstein
2017-06-02 14:04 ` [PATCH 09/17] ovl: allow hashing non upper inodes Amir Goldstein
2017-06-02 14:04 ` [PATCH 10/17] ovl: allow hashing inodes by arbitrary key Amir Goldstein
2017-06-02 14:04 ` [PATCH 11/17] ovl: hash overlay non-dir inodes by copy up origin inode Amir Goldstein
2017-06-05 12:42 ` Amir Goldstein
2017-06-02 14:04 ` [PATCH 12/17] ovl: defer upper dir lock to tempfile link Amir Goldstein
2017-06-02 14:04 ` [PATCH 13/17] ovl: factor out ovl_copy_up_inode() helper Amir Goldstein
2017-06-02 14:04 ` [PATCH 14/17] ovl: generalize ovl_copy_up_locked() using actors Amir Goldstein
2017-06-02 14:04 ` [PATCH 15/17] ovl: generalize ovl_copy_up_one() " Amir Goldstein
2017-06-02 14:04 ` [PATCH 16/17] ovl: implement index dir copy up method Amir Goldstein
2017-06-02 14:04 ` [PATCH 17/17] ovl: handle race of concurrent lower hardlinks copy up Amir Goldstein
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=1496412284-4113-2-git-send-email-amir73il@gmail.com \
--to=amir73il@gmail.com \
--cc=linux-unionfs@vger.kernel.org \
--cc=miklos@szeredi.hu \
/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