From: Cyrill Gorcunov <gorcunov@openvz.org>
To: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Pavel Emelyanov <xemul@parallels.com>,
"Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
Al Viro <viro@zeniv.linux.org.uk>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
Alexey Dobriyan <adobriyan@gmail.com>,
Andrew Morton <akpm@linux-foundation.org>,
James Bottomley <jbottomley@parallels.com>,
Matthew Helsley <matt.helsley@gmail.com>
Subject: Re: [patch 4/8] fs, exportfs: Add export_encode_inode_fh helper
Date: Tue, 21 Aug 2012 17:40:04 +0400 [thread overview]
Message-ID: <20120821134004.GF7670@moon> (raw)
In-Reply-To: <20120821122908.GE9483@fieldses.org>
On Tue, Aug 21, 2012 at 08:29:08AM -0400, J. Bruce Fields wrote:
> > Initial problem -- we don't know what is being watched by an inotify fd.
> >
> > Having a dentry somewhere was the 1st attempt to solve this -- keep a path
> > in inotify and show it when required. It doesn't work since holding a ref on
> > path changes the behavior of watched inode (we cannot rename/unlink/remount
> > it the same way as we could before patching the kernel).
>
> OK. So if you don't mind the fact that there are filesystems with
> inotify support but not filehandle support, then I think generating a
> filehandle early as you describe would work. I guess it's a little more
> memory per watched inode.
So, I thought about something like below, any comments?
---
fs/notify/inotify/inotify.h | 8 +++++++
fs/notify/inotify/inotify_user.c | 41 ++++++++++++++++++++++++++++++++++-----
2 files changed, 44 insertions(+), 5 deletions(-)
Index: linux-2.6.git/fs/notify/inotify/inotify.h
===================================================================
--- linux-2.6.git.orig/fs/notify/inotify/inotify.h
+++ linux-2.6.git/fs/notify/inotify/inotify.h
@@ -1,6 +1,7 @@
#include <linux/fsnotify_backend.h>
#include <linux/inotify.h>
#include <linux/slab.h> /* struct kmem_cache */
+#include <linux/exportfs.h>
extern struct kmem_cache *event_priv_cachep;
@@ -9,9 +10,16 @@ struct inotify_event_private_data {
int wd;
};
+#if defined(CONFIG_PROC_FS) && defined(CONFIG_EXPORTFS) && defined(CONFIG_CHECKPOINT_RESTORE)
+# define INOTIFY_USE_FHANDLE
+#endif
+
struct inotify_inode_mark {
struct fsnotify_mark fsn_mark;
int wd;
+#ifdef INOTIFY_USE_FHANDLE
+ __u8 fhandle[sizeof(struct file_handle) + MAX_HANDLE_SZ];
+#endif
};
extern void inotify_ignored_and_remove_idr(struct fsnotify_mark *fsn_mark,
Index: linux-2.6.git/fs/notify/inotify/inotify_user.c
===================================================================
--- linux-2.6.git.orig/fs/notify/inotify/inotify_user.c
+++ linux-2.6.git/fs/notify/inotify/inotify_user.c
@@ -566,6 +566,32 @@ static void inotify_free_mark(struct fsn
kmem_cache_free(inotify_inode_mark_cachep, i_mark);
}
+#ifdef INOTIFY_USE_FHANDLE
+static int inotify_encode_wd_fhandle(struct inotify_inode_mark *mark, struct dentry *dentry)
+{
+ struct file_handle *fhandle = (struct file_handle *)mark->fhandle;
+ int size, ret;
+
+ BUILD_BUG_ON(sizeof(mark->fhandle) <= sizeof(struct file_handle));
+
+ fhandle->handle_bytes = sizeof(mark->fhandle) - sizeof(struct file_handle);
+ size = fhandle->handle_bytes >> 2;
+
+ ret = exportfs_encode_fh(dentry, (struct fid *)fhandle->f_handle, &size, 0);
+ if ((ret == 255) || (ret == -ENOSPC))
+ return -EOVERFLOW;
+
+ fhandle->handle_type = ret;
+
+ return 0;
+}
+# else
+static int inotify_encode_wd_fhandle(struct inotify_inode_mark *mark, struct dentry *dentry)
+{
+ return 0;
+}
+#endif
+
static int inotify_update_existing_watch(struct fsnotify_group *group,
struct inode *inode,
u32 arg)
@@ -621,10 +647,11 @@ static int inotify_update_existing_watch
}
static int inotify_new_watch(struct fsnotify_group *group,
- struct inode *inode,
+ struct dentry *dentry,
u32 arg)
{
struct inotify_inode_mark *tmp_i_mark;
+ struct inode *inode = dentry->d_inode;
__u32 mask;
int ret;
struct idr *idr = &group->inotify_data.idr;
@@ -647,6 +674,10 @@ static int inotify_new_watch(struct fsno
if (atomic_read(&group->inotify_data.user->inotify_watches) >= inotify_max_user_watches)
goto out_err;
+ ret = inotify_encode_wd_fhandle(tmp_i_mark, dentry);
+ if (ret)
+ goto out_err;
+
ret = inotify_add_to_idr(idr, idr_lock, &group->inotify_data.last_wd,
tmp_i_mark);
if (ret)
@@ -673,16 +704,16 @@ out_err:
return ret;
}
-static int inotify_update_watch(struct fsnotify_group *group, struct inode *inode, u32 arg)
+static int inotify_update_watch(struct fsnotify_group *group, struct dentry *dentry, u32 arg)
{
int ret = 0;
retry:
/* try to update and existing watch with the new arg */
- ret = inotify_update_existing_watch(group, inode, arg);
+ ret = inotify_update_existing_watch(group, dentry->d_inode, arg);
/* no mark present, try to add a new one */
if (ret == -ENOENT)
- ret = inotify_new_watch(group, inode, arg);
+ ret = inotify_new_watch(group, dentry, arg);
/*
* inotify_new_watch could race with another thread which did an
* inotify_new_watch between the update_existing and the add watch
@@ -785,7 +816,7 @@ SYSCALL_DEFINE3(inotify_add_watch, int,
group = filp->private_data;
/* create/update an inode mark */
- ret = inotify_update_watch(group, inode, mask);
+ ret = inotify_update_watch(group, path.dentry, mask);
path_put(&path);
fput_and_out:
fput_light(filp, fput_needed);
next prev parent reply other threads:[~2012-08-21 13:40 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-08-15 9:21 [patch 0/8] procfs, fdinfo updated Cyrill Gorcunov
2012-08-15 9:21 ` [patch 1/8] procfs: Move /proc/pid/fd[info] handling code to fd.[ch] Cyrill Gorcunov
2012-08-15 9:21 ` [patch 2/8] procfs: Convert /proc/pid/fdinfo/ handling routines to seq-file Cyrill Gorcunov
2012-08-15 9:21 ` [patch 3/8] procfs: Add ability to plug in auxiliary fdinfo providers Cyrill Gorcunov
2012-08-15 21:16 ` Al Viro
2012-08-15 21:31 ` Cyrill Gorcunov
2012-08-15 21:29 ` Al Viro
2012-08-15 21:34 ` Cyrill Gorcunov
2012-08-16 10:58 ` Cyrill Gorcunov
2012-08-15 9:21 ` [patch 4/8] fs, exportfs: Add export_encode_inode_fh helper Cyrill Gorcunov
2012-08-15 20:45 ` J. Bruce Fields
2012-08-15 21:02 ` Cyrill Gorcunov
2012-08-15 22:06 ` J. Bruce Fields
2012-08-16 6:24 ` Cyrill Gorcunov
2012-08-16 12:38 ` Cyrill Gorcunov
2012-08-16 12:47 ` J. Bruce Fields
2012-08-16 13:16 ` Cyrill Gorcunov
2012-08-16 14:57 ` Cyrill Gorcunov
2012-08-16 15:05 ` Al Viro
2012-08-16 15:09 ` Cyrill Gorcunov
2012-08-16 13:43 ` Al Viro
2012-08-16 13:47 ` Pavel Emelyanov
2012-08-16 13:50 ` Al Viro
2012-08-16 13:53 ` Pavel Emelyanov
2012-08-16 13:54 ` Cyrill Gorcunov
2012-08-16 14:03 ` James Bottomley
2012-08-16 14:13 ` Pavel Emelyanov
2012-08-16 14:15 ` Cyrill Gorcunov
2012-08-16 14:41 ` Al Viro
2012-08-16 14:48 ` Cyrill Gorcunov
2012-08-16 14:54 ` J. Bruce Fields
2012-08-16 14:55 ` Al Viro
2012-08-16 15:06 ` Pavel Emelyanov
2012-08-16 15:35 ` Cyrill Gorcunov
2012-08-16 15:07 ` Cyrill Gorcunov
2012-08-17 20:58 ` Eric W. Biederman
2012-08-16 13:48 ` Al Viro
2012-08-20 14:19 ` Aneesh Kumar K.V
2012-08-20 16:33 ` Cyrill Gorcunov
2012-08-20 18:32 ` J. Bruce Fields
2012-08-20 19:06 ` Cyrill Gorcunov
2012-08-20 19:32 ` J. Bruce Fields
2012-08-20 19:40 ` Cyrill Gorcunov
2012-08-21 9:18 ` Pavel Emelyanov
2012-08-21 10:42 ` Aneesh Kumar K.V
2012-08-21 10:49 ` Pavel Emelyanov
2012-08-21 10:54 ` Cyrill Gorcunov
2012-08-21 11:09 ` Pavel Emelyanov
2012-08-21 12:11 ` J. Bruce Fields
2012-08-21 12:22 ` Pavel Emelyanov
2012-08-21 12:29 ` J. Bruce Fields
2012-08-21 12:40 ` Pavel Emelyanov
2012-08-21 12:51 ` Boaz Harrosh
2012-08-21 12:59 ` Pavel Emelyanov
2012-08-21 13:08 ` Boaz Harrosh
2012-08-21 13:12 ` Al Viro
2012-08-21 13:40 ` Cyrill Gorcunov [this message]
2012-08-21 12:09 ` J. Bruce Fields
2012-08-21 12:25 ` Pavel Emelyanov
2012-08-22 5:49 ` Aneesh Kumar K.V
2012-08-23 8:06 ` Cyrill Gorcunov
2012-08-23 8:54 ` Marco Stornelli
2012-08-23 9:29 ` Cyrill Gorcunov
2012-08-15 9:21 ` [patch 5/8] fs, notify: Add procfs fdinfo helper v3 Cyrill Gorcunov
2012-08-15 9:21 ` [patch 6/8] fs, eventfd: Add procfs fdinfo helper Cyrill Gorcunov
2012-08-15 9:21 ` [patch 7/8] fs, epoll: Add procfs fdinfo helper v2 Cyrill Gorcunov
2012-08-15 9:21 ` [patch 8/8] fdinfo: Show sigmask for signalfd fd v2 Cyrill Gorcunov
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=20120821134004.GF7670@moon \
--to=gorcunov@openvz.org \
--cc=adobriyan@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=bfields@fieldses.org \
--cc=jbottomley@parallels.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=matt.helsley@gmail.com \
--cc=viro@zeniv.linux.org.uk \
--cc=xemul@parallels.com \
/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;
as well as URLs for NNTP newsgroup(s).