On Mon, Mar 11, 2019 at 11:20:41AM +0000, Jonathan Cameron wrote:

On Wed, 27 Feb 2019 15:50:35 -0700
Keith Busch [off-list ref] wrote:

quoted

+static __init void hmat_register_target_initiators(struct memory_target *target)
+{
+	static DECLARE_BITMAP(p_nodes, MAX_NUMNODES);
+	struct memory_initiator *initiator;
+	unsigned int mem_nid, cpu_nid;
+	struct memory_locality *loc = NULL;
+	u32 best = 0;
+	int i;
+

(upshot of the below is I removed this test :)

quoted

+	if (target->processor_pxm == PXM_INVAL)
+		return;

This doesn't look right.  We check first if it is invalid  and return....

Yeah, Brice mentioned the same bug. I must have been mistakenly
reintroduced that when I rebased to linux-next. I also have a test case
for this and recall it was working at one point. I've got it fixed up
now for the next revision.