Messages in this thread | | | Date | Tue, 3 May 2005 20:28:35 +0530 | From | Dinakar Guniguntala <> | Subject | Re: [RFC PATCH] Dynamic sched domains (v0.5) |
| |
> >On Mon, May 02, 2005 at 07:44:05PM +1000, Nick Piggin wrote: > > What are you protecting against, though? synchroinze_kernel can > sleep, so local_irq_disable is probably the wrong thing to do as well.
Paul, any reason why code marked "####" (fn cpuset_rmdir) is under the dentry lock ??
spin_lock(&cs->dentry->d_lock); parent = cs->parent; #### set_bit(CS_REMOVED, &cs->flags); #### if (is_cpu_exclusive(cs)) update_cpu_domains(cs); list_del(&cs->sibling); /* delete my sibling from parent->children */ if (list_empty(&parent->children)) check_for_release(parent); d = dget(cs->dentry); <---- cs->dentry = NULL; <---- spin_unlock(&d->d_lock);
As far as I can see only the ones marked "<----" should be under the dentry lock, considering the fact that it already holds the cpuset_sem all the while.
I saw that calling update_cpu_domains with the dentry lock held, causes it to oops with preempt turned on. (Scheduling while atomic)
-Dinakar - 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/
| |