mm-commits.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [patch 057/114] ipc/sem: avoid idr tree lookup for interrupted semop
@ 2016-12-14 23:06 akpm
  0 siblings, 0 replies; only message in thread
From: akpm @ 2016-12-14 23:06 UTC (permalink / raw)
  To: torvalds, mm-commits, akpm, dave, dbueso, manfred

From: Davidlohr Bueso <dave@stgolabs.net>
Subject: ipc/sem: avoid idr tree lookup for interrupted semop

We can avoid the idr tree lookup (albeit possibly avoiding
idr_find_fast()) when being awoken in EINTR, as the semid will not change
in this context while blocked.  Use the sma pointer directly and take the
sem_lock, then re-check for RMID races.  We continue to re-check the
queue.status with the lock held such that we can detect situations where
we where are dealing with a spurious wakeup but another task that holds
the sem_lock updated the queue.status while we were spinning for it.  Once
we take the lock it obviously won't change again.

Being the only caller, get rid of sem_obtain_lock() altogether.

Link: http://lkml.kernel.org/r/1478708774-28826-3-git-send-email-dave@stgolabs.net
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Cc: Manfred Spraul <manfred@colorfullife.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 ipc/sem.c |   37 +++++--------------------------------
 1 file changed, 5 insertions(+), 32 deletions(-)

diff -puN ipc/sem.c~ipc-sem-avoid-idr-tree-lookup-for-interrupted-semop ipc/sem.c
--- a/ipc/sem.c~ipc-sem-avoid-idr-tree-lookup-for-interrupted-semop
+++ a/ipc/sem.c
@@ -414,29 +414,6 @@ static inline void sem_unlock(struct sem
  *
  * The caller holds the RCU read lock.
  */
-static inline struct sem_array *sem_obtain_lock(struct ipc_namespace *ns,
-			int id, struct sembuf *sops, int nsops, int *locknum)
-{
-	struct kern_ipc_perm *ipcp;
-	struct sem_array *sma;
-
-	ipcp = ipc_obtain_object_idr(&sem_ids(ns), id);
-	if (IS_ERR(ipcp))
-		return ERR_CAST(ipcp);
-
-	sma = container_of(ipcp, struct sem_array, sem_perm);
-	*locknum = sem_lock(sma, sops, nsops);
-
-	/* ipc_rmid() may have already freed the ID while sem_lock
-	 * was spinning: verify that the structure is still valid
-	 */
-	if (ipc_valid_object(ipcp))
-		return container_of(ipcp, struct sem_array, sem_perm);
-
-	sem_unlock(sma, *locknum);
-	return ERR_PTR(-EINVAL);
-}
-
 static inline struct sem_array *sem_obtain_object(struct ipc_namespace *ns, int id)
 {
 	struct kern_ipc_perm *ipcp = ipc_obtain_object_idr(&sem_ids(ns), id);
@@ -2000,16 +1977,12 @@ SYSCALL_DEFINE4(semtimedop, int, semid,
 		}
 
 		rcu_read_lock();
-		sma = sem_obtain_lock(ns, semid, sops, nsops, &locknum);
-		error = READ_ONCE(queue.status);
+		sem_lock(sma, sops, nsops);
 
-		/*
-		 * Array removed? If yes, leave without sem_unlock().
-		 */
-		if (IS_ERR(sma)) {
-			rcu_read_unlock();
-			goto out_free;
-		}
+		if (!ipc_valid_object(&sma->sem_perm))
+			goto out_unlock_free;
+
+		error = READ_ONCE(queue.status);
 
 		/*
 		 * If queue.status != -EINTR we are woken up by another process.
_

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2016-12-14 23:07 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-12-14 23:06 [patch 057/114] ipc/sem: avoid idr tree lookup for interrupted semop akpm

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).