linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jeff Layton <jlayton@primarydata.com>
To: bfields@fieldses.org
Cc: linux-nfs@vger.kernel.org, hch@infradead.org
Subject: [PATCH 37/40] nfsd: optimize destroy_lockowner cl_lock thrashing
Date: Mon, 21 Jul 2014 11:02:49 -0400	[thread overview]
Message-ID: <1405954972-28904-38-git-send-email-jlayton@primarydata.com> (raw)
In-Reply-To: <1405954972-28904-1-git-send-email-jlayton@primarydata.com>

Reduce the cl_lock trashing in destroy_lockowner. Unhash all of the
lockstateids on the lockowner's list. Put the reference under the lock
and see if it was the last one. If so, then add it to a private list
to be destroyed after we drop the lock.

Signed-off-by: Jeff Layton <jlayton@primarydata.com>
---
 fs/nfsd/nfs4state.c | 50 +++++++++++++++++++++++++++++++++-----------------
 1 file changed, 33 insertions(+), 17 deletions(-)

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 0281b87b77dd..63ee3ebea21c 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -944,14 +944,23 @@ static void put_generic_stateid(struct nfs4_ol_stateid *stp)
 	nfs4_put_stid(&stp->st_stid);
 }
 
-static void release_lock_stateid(struct nfs4_ol_stateid *stp)
+static void unhash_lock_stateid(struct nfs4_ol_stateid *stp)
 {
 	struct nfs4_openowner *oo = openowner(stp->st_openstp->st_stateowner);
 
-	spin_lock(&oo->oo_owner.so_client->cl_lock);
-	list_del(&stp->st_locks);
+	lockdep_assert_held(&oo->oo_owner.so_client->cl_lock);
+
+	list_del_init(&stp->st_locks);
 	unhash_generic_stateid(stp);
 	unhash_stid(&stp->st_stid);
+}
+
+static void release_lock_stateid(struct nfs4_ol_stateid *stp)
+{
+	struct nfs4_openowner *oo = openowner(stp->st_openstp->st_stateowner);
+
+	spin_lock(&oo->oo_owner.so_client->cl_lock);
+	unhash_lock_stateid(stp);
 	spin_unlock(&oo->oo_owner.so_client->cl_lock);
 	put_generic_stateid(stp);
 }
@@ -965,30 +974,37 @@ static void unhash_lockowner_locked(struct nfs4_lockowner *lo)
 	list_del_init(&lo->lo_owner.so_strhash);
 }
 
-static void release_lockowner_stateids(struct nfs4_lockowner *lo)
+static void release_lockowner(struct nfs4_lockowner *lo)
 {
 	struct nfs4_client *clp = lo->lo_owner.so_client;
 	struct nfs4_ol_stateid *stp;
+	struct list_head reaplist;
 
-	lockdep_assert_held(&clp->cl_lock);
+	INIT_LIST_HEAD(&reaplist);
 
+	spin_lock(&clp->cl_lock);
+	unhash_lockowner_locked(lo);
 	while (!list_empty(&lo->lo_owner.so_stateids)) {
 		stp = list_first_entry(&lo->lo_owner.so_stateids,
 				struct nfs4_ol_stateid, st_perstateowner);
-		spin_unlock(&clp->cl_lock);
-		release_lock_stateid(stp);
-		spin_lock(&clp->cl_lock);
+		unhash_lock_stateid(stp);
+		/*
+		 * We now know that no new references can be added to the
+		 * stateid. If ours is the last one, finish the unhashing
+		 * and put it on the list to be reaped.
+		 */
+		if (atomic_dec_and_test(&stp->st_stid.sc_count)) {
+			remove_stid_locked(clp, &stp->st_stid);
+			list_add(&stp->st_locks, &reaplist);
+		}
 	}
-}
-
-static void release_lockowner(struct nfs4_lockowner *lo)
-{
-	struct nfs4_client *clp = lo->lo_owner.so_client;
-
-	spin_lock(&clp->cl_lock);
-	unhash_lockowner_locked(lo);
-	release_lockowner_stateids(lo);
 	spin_unlock(&clp->cl_lock);
+	while (!list_empty(&reaplist)) {
+		stp = list_first_entry(&reaplist, struct nfs4_ol_stateid,
+					st_locks);
+		list_del(&stp->st_locks);
+		stp->st_stid.sc_free(&stp->st_stid);
+	}
 	nfs4_put_stateowner(&lo->lo_owner);
 }
 
-- 
1.9.3


  parent reply	other threads:[~2014-07-21 15:03 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-21 15:02 [PATCH 00/40] nfsd: stateid and stateowner refcounting overhaul Jeff Layton
2014-07-21 15:02 ` [PATCH 01/40] nfsd4: use cl_lock to synchronize all stateid idr calls Jeff Layton
2014-07-27 13:21   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 02/40] nfsd: Add reference counting to the lock and open stateids Jeff Layton
2014-07-27 13:23   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 03/40] nfsd: Add a struct nfs4_file field to struct nfs4_stid Jeff Layton
2014-07-27 13:24   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 04/40] nfsd: Replace nfs4_ol_stateid->st_file with the st_stid.sc_file Jeff Layton
2014-07-27 13:25   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 05/40] nfsd: Ensure atomicity of stateid destruction and idr tree removal Jeff Layton
2014-07-27 13:28   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 06/40] nfsd: Cleanup the freeing of stateids Jeff Layton
2014-07-27 13:35   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 07/40] nfsd: do filp_close in sc_free callback for lock stateids Jeff Layton
2014-07-27 13:37   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 08/40] nfsd: Add locking to protect the state owner lists Jeff Layton
2014-07-27 13:42   ` Christoph Hellwig
2014-07-29 11:42     ` Jeff Layton
2014-07-21 15:02 ` [PATCH 09/40] nfsd: clean up races in lock stateid searching and creation Jeff Layton
2014-07-27 13:44   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 10/40] nfsd: Slight cleanup of find_stateid() Jeff Layton
2014-07-27 13:38   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 11/40] nfsd: ensure atomicity in nfsd4_free_stateid and nfsd4_validate_stateid Jeff Layton
2014-07-27 13:46   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 12/40] nfsd: Add reference counting to lock stateids Jeff Layton
2014-07-21 15:02 ` [PATCH 13/40] nfsd: nfsd4_locku() must reference the lock stateid Jeff Layton
2014-07-21 15:02 ` [PATCH 14/40] nfsd: Ensure that nfs4_open_delegation() references the delegation stateid Jeff Layton
2014-07-21 15:02 ` [PATCH 15/40] nfsd: nfsd4_process_open2() must reference " Jeff Layton
2014-07-21 15:02 ` [PATCH 16/40] nfsd: nfsd4_process_open2() must reference the open stateid Jeff Layton
2014-07-21 15:02 ` [PATCH 17/40] nfsd: Prepare nfsd4_close() for open stateid referencing Jeff Layton
2014-07-21 15:02 ` [PATCH 18/40] nfsd: nfsd4_open_confirm() must reference the open stateid Jeff Layton
2014-07-21 15:02 ` [PATCH 19/40] nfsd: Add reference counting to nfs4_preprocess_confirmed_seqid_op Jeff Layton
2014-07-21 15:02 ` [PATCH 20/40] nfsd: Migrate the stateid reference into nfs4_preprocess_seqid_op Jeff Layton
2014-07-21 15:02 ` [PATCH 21/40] nfsd: Migrate the stateid reference into nfs4_lookup_stateid() Jeff Layton
2014-07-21 15:02 ` [PATCH 22/40] nfsd: Migrate the stateid reference into nfs4_find_stateid_by_type() Jeff Layton
2014-07-21 15:02 ` [PATCH 23/40] nfsd: Add reference counting to state owners Jeff Layton
2014-07-27 13:58   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 24/40] nfsd: Add a mutex to protect the NFSv4.0 open owner replay cache Jeff Layton
2014-07-21 15:02 ` [PATCH 25/40] nfsd: Keep a reference to the open stateid for the NFSv4.0 " Jeff Layton
2014-07-27 13:59   ` Christoph Hellwig
2014-07-21 15:02 ` [PATCH 26/40] nfsd: clean up lockowner refcounting when finding them Jeff Layton
2014-07-21 15:02 ` [PATCH 27/40] nfsd: add an operation for unhashing a stateowner Jeff Layton
2014-07-21 15:02 ` [PATCH 28/40] nfsd: Make lock stateid take a reference to the lockowner Jeff Layton
2014-07-21 15:02 ` [PATCH 29/40] nfsd: clean up refcounting for lockowners Jeff Layton
2014-07-21 15:02 ` [PATCH 30/40] nfsd: make openstateids hold references to their openowners Jeff Layton
2014-07-21 15:02 ` [PATCH 31/40] nfsd: don't allow CLOSE to proceed until refcount on stateid drops Jeff Layton
2014-07-21 15:02 ` [PATCH 32/40] nfsd: Protect adding/removing open state owners using client_lock Jeff Layton
2014-07-21 15:02 ` [PATCH 33/40] nfsd: Protect adding/removing lock " Jeff Layton
2014-07-21 15:02 ` [PATCH 34/40] nfsd: Move the open owner hash table into struct nfs4_client Jeff Layton
2014-07-21 15:02 ` [PATCH 35/40] nfsd: clean up and reorganize release_lockowner Jeff Layton
2014-07-21 15:02 ` [PATCH 36/40] nfsd: add locking to stateowner release Jeff Layton
2014-07-21 15:02 ` Jeff Layton [this message]
2014-07-21 15:02 ` [PATCH 38/40] nfsd: close potential race in nfsd4_free_stateid Jeff Layton
2014-07-21 15:02 ` [PATCH 39/40] nfsd: reduce cl_lock thrashing in release_openowner Jeff Layton
2014-07-21 15:02 ` [PATCH 40/40] nfsd: don't thrash the cl_lock while freeing an open stateid Jeff Layton

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=1405954972-28904-38-git-send-email-jlayton@primarydata.com \
    --to=jlayton@primarydata.com \
    --cc=bfields@fieldses.org \
    --cc=hch@infradead.org \
    --cc=linux-nfs@vger.kernel.org \
    /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).