Return-Path: linux-nfs-owner@vger.kernel.org Received: from fieldses.org ([174.143.236.118]:52677 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756395Ab3FMOui (ORCPT ); Thu, 13 Jun 2013 10:50:38 -0400 Date: Thu, 13 Jun 2013 10:50:09 -0400 From: "J. Bruce Fields" To: Jeff Layton Cc: viro@zeniv.linux.org.uk, matthew@wil.cx, dhowells@redhat.com, sage@inktank.com, smfrench@gmail.com, swhiteho@redhat.com, Trond.Myklebust@netapp.com, akpm@linux-foundation.org, linux-kernel@vger.kernel.org, linux-afs@lists.infradead.org, ceph-devel@vger.kernel.org, linux-cifs@vger.kernel.org, samba-technical@lists.samba.org, cluster-devel@redhat.com, linux-nfs@vger.kernel.org, linux-fsdevel@vger.kernel.org, piastryyy@gmail.com Subject: Re: [PATCH v2 10/14] locks: turn the blocked_list into a hashtable Message-ID: <20130613145009.GB20235@fieldses.org> References: <1370948948-31784-1-git-send-email-jlayton@redhat.com> <1370948948-31784-11-git-send-email-jlayton@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1370948948-31784-11-git-send-email-jlayton@redhat.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Tue, Jun 11, 2013 at 07:09:04AM -0400, Jeff Layton wrote: > Break up the blocked_list into a hashtable, using the fl_owner as a key. > This speeds up searching the hash chains, which is especially significant > for deadlock detection. > > Note that the initial implementation assumes that hashing on fl_owner is > sufficient. In most cases it should be, with the notable exception being > server-side lockd, which compares ownership using a tuple of the > nlm_host and the pid sent in the lock request. So, this may degrade to a > single hash bucket when you only have a single NFS client. That will be > addressed in a later patch. > > The careful observer may note that this patch leaves the file_lock_list > alone. There's much less of a case for turning the file_lock_list into a > hashtable. The only user of that list is the code that generates > /proc/locks, and it always walks the entire list. Makes sense to me, ACK to this and the previous patch.--b. > > Signed-off-by: Jeff Layton > --- > fs/locks.c | 25 ++++++++++++++++++------- > 1 files changed, 18 insertions(+), 7 deletions(-) > > diff --git a/fs/locks.c b/fs/locks.c > index 28959bc..76fb7af 100644 > --- a/fs/locks.c > +++ b/fs/locks.c > @@ -126,6 +126,7 @@ > #include > #include > #include > +#include > > #include > > @@ -153,10 +154,19 @@ int lease_break_time = 45; > #define for_each_lock(inode, lockp) \ > for (lockp = &inode->i_flock; *lockp != NULL; lockp = &(*lockp)->fl_next) > > +/* > + * By breaking up the blocked locks list into a hashtable, we speed up the > + * deadlock detection. > + * > + * FIXME: make this value scale via some heuristic? > + */ > +#define BLOCKED_HASH_BITS 7 > + > +static DEFINE_HASHTABLE(blocked_hash, BLOCKED_HASH_BITS); > + > static HLIST_HEAD(file_lock_list); > -static HLIST_HEAD(blocked_list); > > -/* Protects the two list heads above */ > +/* Protects the file_lock_list and the blocked_hash */ > static DEFINE_SPINLOCK(file_lock_lock); > > static struct kmem_cache *filelock_cache __read_mostly; > @@ -475,13 +485,13 @@ static int posix_same_owner(struct file_lock *fl1, struct file_lock *fl2) > static inline void > locks_insert_global_blocked(struct file_lock *waiter) > { > - hlist_add_head(&waiter->fl_link, &blocked_list); > + hash_add(blocked_hash, &waiter->fl_link, (unsigned long)waiter->fl_owner); > } > > static inline void > __locks_delete_global_blocked(struct file_lock *waiter) > { > - hlist_del_init(&waiter->fl_link); > + hash_del(&waiter->fl_link); > } > > static inline void > @@ -729,7 +739,7 @@ static struct file_lock *what_owner_is_waiting_for(struct file_lock *block_fl) > { > struct file_lock *fl; > > - hlist_for_each_entry(fl, &blocked_list, fl_link) { > + hash_for_each_possible(blocked_hash, fl, fl_link, (unsigned long)block_fl->fl_owner) { > if (posix_same_owner(fl, block_fl)) > return fl->fl_next; > } > @@ -865,7 +875,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str > /* > * New lock request. Walk all POSIX locks and look for conflicts. If > * there are any, either return error or put the request on the > - * blocker's list of waiters and the global blocked_list. > + * blocker's list of waiters and the global blocked_hash. > */ > if (request->fl_type != F_UNLCK) { > for_each_lock(inode, before) { > @@ -2284,13 +2294,14 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl, > > static int locks_show(struct seq_file *f, void *v) > { > + int bkt; > struct file_lock *fl, *bfl; > > fl = hlist_entry(v, struct file_lock, fl_link); > > lock_get_status(f, fl, *((loff_t *)f->private), ""); > > - hlist_for_each_entry(bfl, &blocked_list, fl_link) { > + hash_for_each(blocked_hash, bkt, bfl, fl_link) { > if (bfl->fl_next == fl) > lock_get_status(f, bfl, *((loff_t *)f->private), " ->"); > } > -- > 1.7.1 >