On Wed, 5 Feb 2014, Nathan Zimmer wrote: > > That looks a little problematic, what happens if a nid is being brought > > online and a registered callback does something like allocate resources > > for the arg->status_change_nid and the above two hunks of this patch end > > up racing? > > > > Before, a registered callback would be guaranteed to see either a > > MEMORY_CANCEL_ONLINE or MEMORY_ONLINE after it has already done > > MEMORY_GOING_ONLINE. > > > > With your patch, we could race and see one cpu doing MEMORY_GOING_ONLINE, > > another cpu doing MEMORY_GOING_ONLINE, and then MEMORY_ONLINE and > > MEMORY_CANCEL_ONLINE in either order. > > > > So I think this patch will break most registered callbacks that actually > > depend on lock_memory_hotplug(), it's a coarse lock for that reason. > > Since the argument being passed in is the pfn and size it would be an issue > only if two threads attepted to online the same piece of memory. Right? >
No, I'm referring to registered callbacks that provide a resource for arg->status_change_nid. An example would be the callbacks I added to the slub allocator in slab_memory_callback(). If we are now able to get a racy MEM_GOING_ONLINE -> MEM_GOING_ONLINE -> MEM_ONLINE -> MEM_CANCEL_ONLINE, which is possible with your patch _and_ the node being successfully onlined at the end, then we get a NULL pointer dereference because the kmem_cache_node for each slab cache has been freed. > That seems very unlikely but if it can happen it needs to be protected > against. > The protection for registered memory online or offline callbacks is lock_memory_hotplug() which is eliminated with your patch, the locking for memory_notify() that you're citing is irrelevant. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
