Linux NFS development
 help / color / mirror / Atom feed
From: Anna Schumaker <anna@kernel.org>
To: linux-nfs@vger.kernel.org, trond.myklebust@hammerspace.com
Cc: anna@kernel.org
Subject: [PATCH 1/5] NFS: Clean up locking the nfs_versions list
Date: Tue,  1 Oct 2024 16:33:40 -0400	[thread overview]
Message-ID: <20241001203344.327044-2-anna@kernel.org> (raw)
In-Reply-To: <20241001203344.327044-1-anna@kernel.org>

From: Anna Schumaker <anna.schumaker@oracle.com>

This patch replaces the nfs_version_mutex and nfs_version_lock with a
single RW lock that protects access to the nfs_versions list.

The mutex around request_module() seemed unnecessary to me, and I
couldn't find any other callers using a lock around calls to
request_module() when I looked.

At the same time, I saw fs/filesystems.c using a RW lock to protect
their filesystems list. This seems like a better idea than a spinlock to
me, so I'm also making that change while I'm here.

Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
---
 fs/nfs/client.c | 19 ++++++++-----------
 1 file changed, 8 insertions(+), 11 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index a1d21c4be0ac..15340df117a7 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -55,8 +55,7 @@
 #define NFSDBG_FACILITY		NFSDBG_CLIENT
 
 static DECLARE_WAIT_QUEUE_HEAD(nfs_client_active_wq);
-static DEFINE_SPINLOCK(nfs_version_lock);
-static DEFINE_MUTEX(nfs_version_mutex);
+static DEFINE_RWLOCK(nfs_version_lock);
 static LIST_HEAD(nfs_versions);
 
 /*
@@ -79,16 +78,16 @@ const struct rpc_program nfs_program = {
 static struct nfs_subversion *find_nfs_version(unsigned int version)
 {
 	struct nfs_subversion *nfs;
-	spin_lock(&nfs_version_lock);
+	read_lock(&nfs_version_lock);
 
 	list_for_each_entry(nfs, &nfs_versions, list) {
 		if (nfs->rpc_ops->version == version) {
-			spin_unlock(&nfs_version_lock);
+			read_unlock(&nfs_version_lock);
 			return nfs;
 		}
 	}
 
-	spin_unlock(&nfs_version_lock);
+	read_unlock(&nfs_version_lock);
 	return ERR_PTR(-EPROTONOSUPPORT);
 }
 
@@ -97,10 +96,8 @@ struct nfs_subversion *get_nfs_version(unsigned int version)
 	struct nfs_subversion *nfs = find_nfs_version(version);
 
 	if (IS_ERR(nfs)) {
-		mutex_lock(&nfs_version_mutex);
 		request_module("nfsv%d", version);
 		nfs = find_nfs_version(version);
-		mutex_unlock(&nfs_version_mutex);
 	}
 
 	if (!IS_ERR(nfs) && !try_module_get(nfs->owner))
@@ -115,23 +112,23 @@ void put_nfs_version(struct nfs_subversion *nfs)
 
 void register_nfs_version(struct nfs_subversion *nfs)
 {
-	spin_lock(&nfs_version_lock);
+	write_lock(&nfs_version_lock);
 
 	list_add(&nfs->list, &nfs_versions);
 	nfs_version[nfs->rpc_ops->version] = nfs->rpc_vers;
 
-	spin_unlock(&nfs_version_lock);
+	write_unlock(&nfs_version_lock);
 }
 EXPORT_SYMBOL_GPL(register_nfs_version);
 
 void unregister_nfs_version(struct nfs_subversion *nfs)
 {
-	spin_lock(&nfs_version_lock);
+	write_lock(&nfs_version_lock);
 
 	nfs_version[nfs->rpc_ops->version] = NULL;
 	list_del(&nfs->list);
 
-	spin_unlock(&nfs_version_lock);
+	write_unlock(&nfs_version_lock);
 }
 EXPORT_SYMBOL_GPL(unregister_nfs_version);
 
-- 
2.46.2


  reply	other threads:[~2024-10-01 20:33 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-01 20:33 [PATCH 0/5] NFS: Clean up NFS version module loading Anna Schumaker
2024-10-01 20:33 ` Anna Schumaker [this message]
2024-10-01 20:33 ` [PATCH 2/5] NFS: Convert the NFS module list into an array Anna Schumaker
2024-10-01 20:33 ` [PATCH 3/5] NFS: Rename get_nfs_version() -> find_nfs_version() Anna Schumaker
2024-10-01 20:33 ` [PATCH 4/5] NFS: Clean up find_nfs_version() Anna Schumaker
2024-10-01 20:33 ` [PATCH 5/5] NFS: Implement get_nfs_version() Anna Schumaker

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=20241001203344.327044-2-anna@kernel.org \
    --to=anna@kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=trond.myklebust@hammerspace.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