On Dec 19, 2016 2:58 PM, "Dave Hansen" <dave.hansen@linux.intel.com> wrote:


This boots in a small VM and on a multi-node NUMA system, but has not
been tested widely.


No, this is wrong.

+wait_queue_head_t *bit_waitqueue(void *word, int bit)
+{
+       const int __maybe_unused nid = page_to_nid(virt_to_page(word));
+
+       return __bit_waitqueue(word, bit, nid);


No can do. Part of the problem with the old coffee was that it did that
virt_to_page() crud. That doesn't work with the virtually mapped stack.

So bit_waitqueue() must not do the page lookup.

Only [un]lock_page() that already has a page can do the NID thing.

OK?

     Linus