From: Jeff Layton <jlayton@kernel.org>
To: NeilBrown <neil@brown.name>,
Olga Kornievskaia <okorniev@redhat.com>,
Dai Ngo <Dai.Ngo@oracle.com>, Tom Talpey <tom@talpey.com>,
Trond Myklebust <trondmy@kernel.org>,
Anna Schumaker <anna@kernel.org>,
Jonathan Corbet <corbet@lwn.net>,
Shuah Khan <skhan@linuxfoundation.org>,
Chuck Lever <cel@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>,
Alexander Aring <alex.aring@gmail.com>,
Amir Goldstein <amir73il@gmail.com>, Jan Kara <jack@suse.cz>,
Alexander Viro <viro@zeniv.linux.org.uk>,
Christian Brauner <brauner@kernel.org>,
Calum Mackay <calum.mackay@oracle.com>,
linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org,
linux-nfs@vger.kernel.org, Jeff Layton <jlayton@kernel.org>
Subject: [PATCH v6 05/20] nfsd: update the fsnotify mark when setting or removing a dir delegation
Date: Thu, 11 Jun 2026 13:50:11 -0400 [thread overview]
Message-ID: <20260611-dir-deleg-v6-5-4c45080e5f3f@kernel.org> (raw)
In-Reply-To: <20260611-dir-deleg-v6-0-4c45080e5f3f@kernel.org>
Add a new helper function that will update the mask on the nfsd_file's
fsnotify_mark to be a union of all current directory delegations on an
inode.
Call that when directory delegations are added or removed, since that
can change what fsnotify events nfsd requires from the VFS layer.
The fsnotify_mark is shared by every nfsd_file open on the inode, so
concurrent delegation adds and removes on the same directory can run
nfsd_fsnotify_recalc_mask() in parallel. Because it reads the lease
state and updates the mark in two separate locked sections, a recalc
working from a stale snapshot of the lease list could clobber a
concurrent update and leave the mark missing required events. Add an
nfm_recalc_mutex to the nfsd_file_mark and hold it across the recalc to
serialize callers.
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/nfsd/filecache.c | 52 ++++++++++++++++++++++++++++++++++++++++++++++++++++
fs/nfsd/filecache.h | 3 +++
fs/nfsd/nfs4state.c | 5 +++--
3 files changed, 58 insertions(+), 2 deletions(-)
diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c
index 1ea2bfd51825..c5f2c5768324 100644
--- a/fs/nfsd/filecache.c
+++ b/fs/nfsd/filecache.c
@@ -192,6 +192,7 @@ nfsd_file_mark_find_or_create(struct inode *inode)
fsnotify_init_mark(&new->nfm_mark, nfsd_file_fsnotify_group);
new->nfm_mark.mask = FS_ATTRIB|FS_DELETE_SELF;
refcount_set(&new->nfm_ref, 1);
+ mutex_init(&new->nfm_recalc_mutex);
err = fsnotify_add_inode_mark(&new->nfm_mark, inode, 0);
@@ -1473,3 +1474,54 @@ int nfsd_file_cache_stats_show(struct seq_file *m, void *v)
seq_printf(m, "mean age (ms): -\n");
return 0;
}
+
+/**
+ * nfsd_fsnotify_recalc_mask - recalculate the fsnotify mask for a nfsd_file
+ * @nf: nfsd_file to recalculate the mask on
+ *
+ * When a directory nfsd_file has a delegation added or removed, that may
+ * change the events that nfsd requires from the VFS layer. This function
+ * recalculates the fsnotify mask based on the leases present.
+ */
+void nfsd_fsnotify_recalc_mask(struct nfsd_file *nf)
+{
+ struct inode *inode = file_inode(nf->nf_file);
+ u32 lease_mask, set = 0, clear = 0;
+ struct fsnotify_mark *mark;
+
+ /* This is only needed when adding or removing dir delegs */
+ if (!S_ISDIR(inode->i_mode) || !nf->nf_mark)
+ return;
+
+ mark = &nf->nf_mark->nfm_mark;
+
+ /*
+ * The mark is shared by every nfsd_file on this inode, so concurrent
+ * delegation add/remove on the same directory can recalc it in
+ * parallel. Serialize the read of the lease state and the update of
+ * the mark so that a recalc working from a stale snapshot of the
+ * lease list can't clobber a concurrent recalc's update.
+ */
+ mutex_lock(&nf->nf_mark->nfm_recalc_mutex);
+
+ /* Set up notifications for any ignored delegation events */
+ lease_mask = inode_lease_ignore_mask(inode);
+
+ if (lease_mask & FL_IGN_DIR_CREATE)
+ set |= FS_CREATE | FS_MOVED_TO;
+ else
+ clear |= FS_CREATE | FS_MOVED_TO;
+
+ if (lease_mask & FL_IGN_DIR_DELETE)
+ set |= FS_DELETE | FS_MOVED_FROM;
+ else
+ clear |= FS_DELETE | FS_MOVED_FROM;
+
+ if (lease_mask & FL_IGN_DIR_RENAME)
+ set |= FS_RENAME;
+ else
+ clear |= FS_RENAME;
+
+ fsnotify_modify_mark_mask(mark, set, clear);
+ mutex_unlock(&nf->nf_mark->nfm_recalc_mutex);
+}
diff --git a/fs/nfsd/filecache.h b/fs/nfsd/filecache.h
index 683b6437cacc..b224902b438d 100644
--- a/fs/nfsd/filecache.h
+++ b/fs/nfsd/filecache.h
@@ -26,6 +26,8 @@
struct nfsd_file_mark {
struct fsnotify_mark nfm_mark;
refcount_t nfm_ref;
+ /* serializes nfsd_fsnotify_recalc_mask() against itself */
+ struct mutex nfm_recalc_mutex;
};
/*
@@ -86,4 +88,5 @@ __be32 nfsd_file_acquire_local(struct net *net, struct svc_cred *cred,
__be32 nfsd_file_acquire_dir(struct svc_rqst *rqstp, struct svc_fh *fhp,
struct nfsd_file **pnf);
int nfsd_file_cache_stats_show(struct seq_file *m, void *v);
+void nfsd_fsnotify_recalc_mask(struct nfsd_file *nf);
#endif /* _FS_NFSD_FILECACHE_H */
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index ae8505747dc2..0cbb37f73ee7 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -1255,6 +1255,7 @@ static void nfs4_unlock_deleg_lease(struct nfs4_delegation *dp)
nfsd4_finalize_deleg_timestamps(dp, nf->nf_file);
kernel_setlease(nf->nf_file, F_UNLCK, NULL, (void **)&dp);
+ nfsd_fsnotify_recalc_mask(nf);
put_deleg_file(fp);
}
@@ -9725,8 +9726,7 @@ nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp, struct dentry *dentry,
* @nf: nfsd_file opened on the directory
*
* Given a GET_DIR_DELEGATION request @gdd, attempt to acquire a delegation
- * on the directory to which @nf refers. Note that this does not set up any
- * sort of async notifications for the delegation.
+ * on the directory to which @nf refers.
*/
struct nfs4_delegation *
nfsd_get_dir_deleg(struct nfsd4_compound_state *cstate,
@@ -9816,6 +9816,7 @@ nfsd_get_dir_deleg(struct nfsd4_compound_state *cstate,
if (!status) {
put_nfs4_file(fp);
+ nfsd_fsnotify_recalc_mask(nf);
return dp;
}
--
2.54.0
next prev parent reply other threads:[~2026-06-11 17:50 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-11 17:50 [PATCH v6 00/20] nfsd: add support for CB_NOTIFY callbacks in directory delegations Jeff Layton
2026-06-11 17:50 ` [PATCH v6 01/20] nfsd: check fl_lmops in nfsd_breaker_owns_lease() Jeff Layton
2026-06-11 17:50 ` [PATCH v6 02/20] nfsd: add protocol support for CB_NOTIFY Jeff Layton
2026-06-11 21:33 ` Chuck Lever
2026-06-11 17:50 ` [PATCH v6 03/20] nfs_common: add new NOTIFY4_* flags proposed in RFC8881bis Jeff Layton
2026-06-11 17:50 ` [PATCH v6 04/20] nfsd: allow nfsd to get a dir lease with an ignore mask Jeff Layton
2026-06-11 17:50 ` Jeff Layton [this message]
2026-06-11 17:50 ` [PATCH v6 06/20] nfsd: make nfsd4_callback_ops->prepare operation bool return Jeff Layton
2026-06-11 17:50 ` [PATCH v6 07/20] nfsd: add callback encoding and decoding linkages for CB_NOTIFY Jeff Layton
2026-06-11 17:50 ` [PATCH v6 08/20] nfsd: use RCU to protect fi_deleg_file Jeff Layton
2026-06-11 17:50 ` [PATCH v6 09/20] nfsd: add data structures for handling CB_NOTIFY Jeff Layton
2026-06-11 17:50 ` [PATCH v6 10/20] nfsd: add notification handlers for dir events Jeff Layton
2026-06-12 17:51 ` Chuck Lever
2026-06-12 18:36 ` Jeff Layton
2026-06-11 17:50 ` [PATCH v6 11/20] nfsd: apply the notify mask to the delegation when requested Jeff Layton
2026-06-12 17:57 ` Chuck Lever
2026-06-11 17:50 ` [PATCH v6 12/20] nfsd: add helper to marshal a fattr4 from completed args Jeff Layton
2026-06-11 17:50 ` [PATCH v6 13/20] nfsd: allow nfsd4_encode_fattr4_change() to work with no export Jeff Layton
2026-06-11 17:50 ` [PATCH v6 14/20] nfsd: send basic file attributes in CB_NOTIFY Jeff Layton
2026-06-11 17:50 ` [PATCH v6 15/20] nfsd: allow encoding a filehandle into fattr4 without a svc_fh Jeff Layton
2026-06-12 18:03 ` Chuck Lever
2026-06-11 17:50 ` [PATCH v6 16/20] nfsd: add a fi_connectable flag to struct nfs4_file Jeff Layton
2026-06-12 18:06 ` Chuck Lever
2026-06-11 17:50 ` [PATCH v6 17/20] nfsd: add the filehandle to returned attributes in CB_NOTIFY Jeff Layton
2026-06-12 18:08 ` Chuck Lever
2026-06-11 17:50 ` [PATCH v6 18/20] nfsd: properly track requested child attributes Jeff Layton
2026-06-12 18:10 ` Chuck Lever
2026-06-11 17:50 ` [PATCH v6 19/20] nfsd: track requested dir attributes Jeff Layton
2026-06-12 18:13 ` Chuck Lever
2026-06-11 17:50 ` [PATCH v6 20/20] nfsd: add support to CB_NOTIFY for dir attribute changes Jeff Layton
2026-06-12 18:21 ` Chuck Lever
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=20260611-dir-deleg-v6-5-4c45080e5f3f@kernel.org \
--to=jlayton@kernel.org \
--cc=Dai.Ngo@oracle.com \
--cc=alex.aring@gmail.com \
--cc=amir73il@gmail.com \
--cc=anna@kernel.org \
--cc=brauner@kernel.org \
--cc=calum.mackay@oracle.com \
--cc=cel@kernel.org \
--cc=corbet@lwn.net \
--cc=jack@suse.cz \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=neil@brown.name \
--cc=okorniev@redhat.com \
--cc=rostedt@goodmis.org \
--cc=skhan@linuxfoundation.org \
--cc=tom@talpey.com \
--cc=trondmy@kernel.org \
--cc=viro@zeniv.linux.org.uk \
/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