Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753170Ab2HUNkQ (ORCPT ); Tue, 21 Aug 2012 09:40:16 -0400 Received: from mail-lpp01m010-f46.google.com ([209.85.215.46]:40877 "EHLO mail-lpp01m010-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752976Ab2HUNkJ (ORCPT ); Tue, 21 Aug 2012 09:40:09 -0400 Date: Tue, 21 Aug 2012 17:40:04 +0400 From: Cyrill Gorcunov To: "J. Bruce Fields" Cc: Pavel Emelyanov , "Aneesh Kumar K.V" , Al Viro , "linux-kernel@vger.kernel.org" , "linux-fsdevel@vger.kernel.org" , Alexey Dobriyan , Andrew Morton , James Bottomley , Matthew Helsley Subject: Re: [patch 4/8] fs, exportfs: Add export_encode_inode_fh helper Message-ID: <20120821134004.GF7670@moon> References: <20120820190606.GE27443@moon> <20120820193204.GD5779@fieldses.org> <50335261.5090504@parallels.com> <87wr0sle4v.fsf@skywalker.in.ibm.com> <503367CB.9080609@parallels.com> <20120821105424.GA7670@moon> <50336C51.60501@parallels.com> <20120821121155.GC9483@fieldses.org> <50337D87.90607@parallels.com> <20120821122908.GE9483@fieldses.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120821122908.GE9483@fieldses.org> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5027 Lines: 143 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 #include #include /* struct kmem_cache */ +#include 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); -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/