Re: recalc_sigpending() / recalc_sigpending_tsk() ?

From: David Woodhouse (dwmw2@infradead.org)
Date: Fri Mar 01 2002 - 12:05:06 EST

torvalds@transmeta.com said:
> So the basic rule should be: drivers etc should not ever have to
> touch "sigmask_lock", because they simply should never even _know_
> about things like that. Agreed?

Absolutely.

> So I would suggest solving this problem by just adding something like
> /* Block all signals except for mask */
> void sigallow(unsigned long mask)
> {
> spin_lock_irq(&current->sigmask_lock);
> siginitsetinv(current->blocked, mask);
> recalc_sigpending();
> spin_unlock_irq(&current->sigmask_lock);
> }

Now that I like. Especially if coupled with the equivalent for dequeuing a
signal with implicit locking too, so I really can drop all references to
current->sigmask_lock -- the JFFS2 garbage collection thread dequeues
signals and deals as you might expect with SIGSTOP/SIGKILL:

        for (;;) {
                spin_lock_irq(&current->sigmask_lock);
                siginitsetinv (&current->blocked, sigmask(SIGHUP) | sigmask(SIGKILL) | sigmask(SIGSTOP) | sigmask(SIGCONT));
                recalc_sigpending();
                spin_unlock_irq(&current->sigmask_lock);

                if (!thread_should_wake(c)) {
                        set_current_state (TASK_INTERRUPTIBLE);
                        D1(printk(KERN_DEBUG "jffs2_garbage_collect_thread sleeping...\n"));
                        /* Yes, there's a race here; we checked thread_should_wake() before
                           setting current->state to TASK_INTERRUPTIBLE. But it doesn't
                           matter - We don't care if we miss a wakeup, because the GC thread
                           is only an optimisation anyway. */
                        schedule();
                }
                
                cond_resched();

                /* Put_super will send a SIGKILL and then wait on the sem.
                 */
                while (signal_pending(current)) {
                        siginfo_t info;
                        unsigned long signr;

                        spin_lock_irq(&current->sigmask_lock);
                        signr = dequeue_signal(&current->blocked, &info);
                        spin_unlock_irq(&current->sigmask_lock);

                        switch(signr) {
                        case SIGSTOP:
                                D1(printk(KERN_DEBUG "jffs2_garbage_collect_thread(): SIGSTOP received.\n"));
                                set_current_state(TASK_STOPPED);
                                schedule();
                                break;

                        case SIGKILL:
                                D1(printk(KERN_DEBUG "jffs2_garbage_collect_thread(): SIGKILL received.\n"));
                                spin_lock_bh(&c->erase_completion_lock);
                                c->gc_task = NULL;
                                spin_unlock_bh(&c->erase_completion_lock);
                                complete_and_exit(&c->gc_thread_exit, 0);

                        case SIGHUP:
                                D1(printk(KERN_DEBUG "jffs2_garbage_collect_thread(): SIGHUP received.\n"));
                                break;
                        default:
                                D1(printk(KERN_DEBUG "jffs2_garbage_collect_thread(): signal %ld received\n", signr));

                        }
                }
                /* We don't want SIGHUP to interrupt us. STOP and KILL are OK though. */
                spin_lock_irq(&current->sigmask_lock);
                siginitsetinv (&current->blocked, sigmask(SIGKILL) | sigmask(SIGSTOP) | sigmask(SIGCONT));
                recalc_sigpending();
                spin_unlock_irq(&current->sigmask_lock);

                D1(printk(KERN_DEBUG "jffs2_garbage_collect_thread(): pass\n"));
                jffs2_garbage_collect_pass(c);
        } 

--
dwmw2

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

This archive was generated by hypermail 2b29 : Thu Mar 07 2002 - 21:00:17 EST