ipc/sem.c: rename try_atomic_semop() to perform_atomic_semop(), docu update
authorManfred Spraul <manfred@colorfullife.com>
Mon, 8 Jul 2013 23:01:26 +0000 (16:01 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 9 Jul 2013 17:33:28 +0000 (10:33 -0700)
Cleanup: Some minor points that I noticed while writing the previous
patches

1) The name try_atomic_semop() is misleading: The function performs the
   operation (if it is possible).

2) Some documentation updates.

No real code change, a rename and documentation changes.

Signed-off-by: Manfred Spraul <manfred@colorfullife.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Davidlohr Bueso <davidlohr.bueso@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
ipc/sem.c

index 51352e1bfff912c6b414be3b494995c5e0e2556d..41088899783d4106140333014a722da531494838 100644 (file)
--- a/ipc/sem.c
+++ b/ipc/sem.c
@@ -154,12 +154,15 @@ static int sysvipc_sem_proc_show(struct seq_file *s, void *it);
 #define SEMOPM_FAST    64  /* ~ 372 bytes on stack */
 
 /*
- * linked list protection:
+ * Locking:
  *     sem_undo.id_next,
+ *     sem_array.complex_count,
  *     sem_array.pending{_alter,_cont},
- *     sem_array.sem_undo: sem_lock() for read/write
+ *     sem_array.sem_undo: global sem_lock() for read/write
  *     sem_undo.proc_next: only "current" is allowed to read/write that field.
  *     
+ *     sem_array.sem_base[i].pending_{const,alter}:
+ *             global or semaphore sem_lock() for read/write
  */
 
 #define sc_semmsl      sem_ctls[0]
@@ -536,12 +539,19 @@ SYSCALL_DEFINE3(semget, key_t, key, int, nsems, int, semflg)
        return ipcget(ns, &sem_ids(ns), &sem_ops, &sem_params);
 }
 
-/*
- * Determine whether a sequence of semaphore operations would succeed
- * all at once. Return 0 if yes, 1 if need to sleep, else return error code.
+/** perform_atomic_semop - Perform (if possible) a semaphore operation
+ * @sma: semaphore array
+ * @sops: array with operations that should be checked
+ * @nsems: number of sops
+ * @un: undo array
+ * @pid: pid that did the change
+ *
+ * Returns 0 if the operation was possible.
+ * Returns 1 if the operation is impossible, the caller must sleep.
+ * Negative values are error codes.
  */
 
-static int try_atomic_semop (struct sem_array * sma, struct sembuf * sops,
+static int perform_atomic_semop(struct sem_array *sma, struct sembuf *sops,
                             int nsops, struct sem_undo *un, int pid)
 {
        int result, sem_op;
@@ -724,8 +734,8 @@ static int wake_const_ops(struct sem_array *sma, int semnum,
                q = container_of(walk, struct sem_queue, list);
                walk = walk->next;
 
-               error = try_atomic_semop(sma, q->sops, q->nsops,
-                                               q->undo, q->pid);
+               error = perform_atomic_semop(sma, q->sops, q->nsops,
+                                                q->undo, q->pid);
 
                if (error <= 0) {
                        /* operation completed, remove from queue & wakeup */
@@ -838,7 +848,7 @@ again:
                if (semnum != -1 && sma->sem_base[semnum].semval == 0)
                        break;
 
-               error = try_atomic_semop(sma, q->sops, q->nsops,
+               error = perform_atomic_semop(sma, q->sops, q->nsops,
                                         q->undo, q->pid);
 
                /* Does q->sleeper still need to sleep? */
@@ -1686,7 +1696,6 @@ static int get_queue_result(struct sem_queue *q)
        return error;
 }
 
-
 SYSCALL_DEFINE4(semtimedop, int, semid, struct sembuf __user *, tsops,
                unsigned, nsops, const struct timespec __user *, timeout)
 {
@@ -1784,7 +1793,8 @@ SYSCALL_DEFINE4(semtimedop, int, semid, struct sembuf __user *, tsops,
        if (un && un->semid == -1)
                goto out_unlock_free;
 
-       error = try_atomic_semop (sma, sops, nsops, un, task_tgid_vnr(current));
+       error = perform_atomic_semop(sma, sops, nsops, un,
+                                       task_tgid_vnr(current));
        if (error <= 0) {
                if (alter && error == 0)
                        do_smart_update(sma, sops, nsops, 1, &tasks);