mm/mempolicy: use unified 'nodes' for bind/interleave/prefer policies

Current structure 'mempolicy' uses a union to store the node info for
bind/interleave/perfer policies.

union {
short preferred_node; /* preferred */
nodemask_t nodes; /* interleave/bind */
/* undefined for default */
} v;

Since preferred node can also be represented by a nodemask_t with only ont
bit set, unify these policies with using one nodemask_t 'nodes', which can
remove a union, simplify the code and make it easier to support future's
new policy's node info.

Link: https://lore.kernel.org/r/[e-mail]
Link: https://lkml.kernel.org/r/[e-mail]
Co-developed-by: Feng Tang <[e-mail]>
Signed-off-by: Ben Widawsky <[e-mail]>
Signed-off-by: Feng Tang <[e-mail]>
Cc: Michal Hocko <[e-mail]>
Cc: David Rientjes <[e-mail]>
Cc: Dave Hansen <[e-mail]>
Cc: Andrea Arcangeli <[e-mail]>
Cc: Mel Gorman <[e-mail]>
Cc: Mike Kravetz <[e-mail]>
Cc: Vlastimil Babka <[e-mail]>
Cc: Andi Kleen <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>
Signed-off-by: Linus Torvalds <[e-mail]>

include/linux/mempolicy.h

···
 	atomic_t refcnt;
 	unsigned short mode; 	/* See MPOL_* above */
 	unsigned short flags;	/* See set_mempolicy() MPOL_F_* above */
-	union {
-		short 		 preferred_node; /* preferred */
-		nodemask_t	 nodes;		/* interleave/bind */
-		/* undefined for default */
-	} v;
+	nodemask_t nodes;	/* interleave/bind/perfer */
+
 	union {
 		nodemask_t cpuset_mems_allowed;	/* relative to these nodes */
 		nodemask_t user_nodemask;	/* nodemask passed by user */

mm/mempolicy.c

···
 {
 	if (nodes_empty(*nodes))
 		return -EINVAL;
-	pol->v.nodes = *nodes;
+	pol->nodes = *nodes;
 	return 0;
 }
 
···
 {
 	if (nodes_empty(*nodes))
 		return -EINVAL;
-	pol->v.preferred_node = first_node(*nodes);
+
+	nodes_clear(pol->nodes);
+	node_set(first_node(*nodes), pol->nodes);
 	return 0;
 }
 
···
 {
 	if (nodes_empty(*nodes))
 		return -EINVAL;
-	pol->v.nodes = *nodes;
+	pol->nodes = *nodes;
 	return 0;
 }
 
···
 	else if (pol->flags & MPOL_F_RELATIVE_NODES)
 		mpol_relative_nodemask(&tmp, &pol->w.user_nodemask, nodes);
 	else {
-		nodes_remap(tmp, pol->v.nodes, pol->w.cpuset_mems_allowed,
+		nodes_remap(tmp, pol->nodes, pol->w.cpuset_mems_allowed,
 								*nodes);
 		pol->w.cpuset_mems_allowed = *nodes;
 	}
···
 	if (nodes_empty(tmp))
 		tmp = *nodes;
 
-	pol->v.nodes = tmp;
+	pol->nodes = tmp;
 }
 
 static void mpol_rebind_preferred(struct mempolicy *pol,
···
 	switch (p->mode) {
 	case MPOL_BIND:
 	case MPOL_INTERLEAVE:
-		*nodes = p->v.nodes;
+	case MPOL_PREFERRED:
+		*nodes = p->nodes;
 		break;
 	case MPOL_LOCAL:
 		/* return empty node mask for local allocation */
-		break;
-
-	case MPOL_PREFERRED:
-		node_set(p->v.preferred_node, *nodes);
 		break;
 	default:
 		BUG();
···
 			*policy = err;
 		} else if (pol == current->mempolicy &&
 				pol->mode == MPOL_INTERLEAVE) {
-			*policy = next_node_in(current->il_prev, pol->v.nodes);
+			*policy = next_node_in(current->il_prev, pol->nodes);
 		} else {
 			err = -EINVAL;
 			goto out;
···
 	BUG_ON(dynamic_policy_zone == ZONE_MOVABLE);
 
 	/*
-	 * if policy->v.nodes has movable memory only,
+	 * if policy->nodes has movable memory only,
 	 * we apply policy when gfp_zone(gfp) = ZONE_MOVABLE only.
 	 *
-	 * policy->v.nodes is intersect with node_states[N_MEMORY].
+	 * policy->nodes is intersect with node_states[N_MEMORY].
 	 * so if the following test fails, it implies
-	 * policy->v.nodes has movable memory only.
+	 * policy->nodes has movable memory only.
 	 */
-	if (!nodes_intersects(policy->v.nodes, node_states[N_HIGH_MEMORY]))
+	if (!nodes_intersects(policy->nodes, node_states[N_HIGH_MEMORY]))
 		dynamic_policy_zone = ZONE_MOVABLE;
 
 	return zone >= dynamic_policy_zone;
···
 	/* Lower zones don't get a nodemask applied for MPOL_BIND */
 	if (unlikely(policy->mode == MPOL_BIND) &&
 			apply_policy_zone(policy, gfp_zone(gfp)) &&
-			cpuset_nodemask_valid_mems_allowed(&policy->v.nodes))
-		return &policy->v.nodes;
+			cpuset_nodemask_valid_mems_allowed(&policy->nodes))
+		return &policy->nodes;
 
 	return NULL;
 }
···
 static int policy_node(gfp_t gfp, struct mempolicy *policy, int nd)
 {
 	if (policy->mode == MPOL_PREFERRED) {
-		nd = policy->v.preferred_node;
+		nd = first_node(policy->nodes);
 	} else {
 		/*
 		 * __GFP_THISNODE shouldn't even be used with the bind policy
···
 	unsigned next;
 	struct task_struct *me = current;
 
-	next = next_node_in(me->il_prev, policy->v.nodes);
+	next = next_node_in(me->il_prev, policy->nodes);
 	if (next < MAX_NUMNODES)
 		me->il_prev = next;
 	return next;
···
 
 	switch (policy->mode) {
 	case MPOL_PREFERRED:
-		return policy->v.preferred_node;
+		return first_node(policy->nodes);
 
 	case MPOL_INTERLEAVE:
 		return interleave_nodes(policy);
···
 		enum zone_type highest_zoneidx = gfp_zone(GFP_KERNEL);
 		zonelist = &NODE_DATA(node)->node_zonelists[ZONELIST_FALLBACK];
 		z = first_zones_zonelist(zonelist, highest_zoneidx,
-							&policy->v.nodes);
+							&policy->nodes);
 		return z->zone ? zone_to_nid(z->zone) : node;
 	}
 	case MPOL_LOCAL:
···
 
 /*
  * Do static interleaving for a VMA with known offset @n.  Returns the n'th
- * node in pol->v.nodes (starting from n=0), wrapping around if n exceeds the
+ * node in pol->nodes (starting from n=0), wrapping around if n exceeds the
  * number of present nodes.
  */
 static unsigned offset_il_node(struct mempolicy *pol, unsigned long n)
 {
-	unsigned nnodes = nodes_weight(pol->v.nodes);
+	unsigned nnodes = nodes_weight(pol->nodes);
 	unsigned target;
 	int i;
 	int nid;
···
 	if (!nnodes)
 		return numa_node_id();
 	target = (unsigned int)n % nnodes;
-	nid = first_node(pol->v.nodes);
+	nid = first_node(pol->nodes);
 	for (i = 0; i < target; i++)
-		nid = next_node(nid, pol->v.nodes);
+		nid = next_node(nid, pol->nodes);
 	return nid;
 }
 
···
 	} else {
 		nid = policy_node(gfp_flags, *mpol, numa_node_id());
 		if ((*mpol)->mode == MPOL_BIND)
-			*nodemask = &(*mpol)->v.nodes;
+			*nodemask = &(*mpol)->nodes;
 	}
 	return nid;
 }
···
 bool init_nodemask_of_mempolicy(nodemask_t *mask)
 {
 	struct mempolicy *mempolicy;
-	int nid;
 
 	if (!(mask && current->mempolicy))
 		return false;
···
 	mempolicy = current->mempolicy;
 	switch (mempolicy->mode) {
 	case MPOL_PREFERRED:
-		nid = mempolicy->v.preferred_node;
-		init_nodemask_of_node(mask, nid);
-		break;
-
 	case MPOL_BIND:
 	case MPOL_INTERLEAVE:
-		*mask = mempolicy->v.nodes;
+		*mask = mempolicy->nodes;
 		break;
 
 	case MPOL_LOCAL:
-		nid = numa_node_id();
-		init_nodemask_of_node(mask, nid);
+		init_nodemask_of_node(mask, numa_node_id());
 		break;
 
 	default:
···
 	task_lock(tsk);
 	mempolicy = tsk->mempolicy;
 	if (mempolicy && mempolicy->mode == MPOL_BIND)
-		ret = nodes_intersects(mempolicy->v.nodes, *mask);
+		ret = nodes_intersects(mempolicy->nodes, *mask);
 	task_unlock(tsk);
 
 	return ret;
···
 		 * node in its nodemask, we allocate the standard way.
 		 */
 		if (pol->mode == MPOL_PREFERRED)
-			hpage_node = pol->v.preferred_node;
+			hpage_node = first_node(pol->nodes);
 
 		nmask = policy_nodemask(gfp, pol);
 		if (!nmask || node_isset(hpage_node, *nmask)) {
···
 	switch (a->mode) {
 	case MPOL_BIND:
 	case MPOL_INTERLEAVE:
-		return !!nodes_equal(a->v.nodes, b->v.nodes);
 	case MPOL_PREFERRED:
-		return a->v.preferred_node == b->v.preferred_node;
+		return !!nodes_equal(a->nodes, b->nodes);
 	case MPOL_LOCAL:
 		return true;
 	default:
···
 		break;
 
 	case MPOL_PREFERRED:
-		polnid = pol->v.preferred_node;
+		polnid = first_node(pol->nodes);
 		break;
 
 	case MPOL_LOCAL:
···
 	case MPOL_BIND:
 		/* Optimize placement among multiple nodes via NUMA balancing */
 		if (pol->flags & MPOL_F_MORON) {
-			if (node_isset(thisnid, pol->v.nodes))
+			if (node_isset(thisnid, pol->nodes))
 				break;
 			goto out;
 		}
···
 		 * else select nearest allowed node, if any.
 		 * If no allowed nodes, use current [!misplaced].
 		 */
-		if (node_isset(curnid, pol->v.nodes))
+		if (node_isset(curnid, pol->nodes))
 			goto out;
 		z = first_zones_zonelist(
 				node_zonelist(numa_node_id(), GFP_HIGHUSER),
 				gfp_zone(GFP_HIGHUSER),
-				&pol->v.nodes);
+				&pol->nodes);
 		polnid = zone_to_nid(z->zone);
 		break;
 
···
 		 vma->vm_pgoff,
 		 sz, npol ? npol->mode : -1,
 		 npol ? npol->flags : -1,
-		 npol ? nodes_addr(npol->v.nodes)[0] : NUMA_NO_NODE);
+		 npol ? nodes_addr(npol->nodes)[0] : NUMA_NO_NODE);
 
 	if (npol) {
 		new = sp_alloc(vma->vm_pgoff, vma->vm_pgoff + sz, npol);
···
 			.refcnt = ATOMIC_INIT(1),
 			.mode = MPOL_PREFERRED,
 			.flags = MPOL_F_MOF | MPOL_F_MORON,
-			.v = { .preferred_node = nid, },
+			.nodes = nodemask_of_node(nid),
 		};
 	}
 
···
 	 * Save nodes for mpol_to_str() to show the tmpfs mount options
 	 * for /proc/mounts, /proc/pid/mounts and /proc/pid/mountinfo.
 	 */
-	if (mode != MPOL_PREFERRED)
-		new->v.nodes = nodes;
-	else if (nodelist)
-		new->v.preferred_node = first_node(nodes);
-	else
+	if (mode != MPOL_PREFERRED) {
+		new->nodes = nodes;
+	} else if (nodelist) {
+		nodes_clear(new->nodes);
+		node_set(first_node(nodes), new->nodes);
+	} else {
 		new->mode = MPOL_LOCAL;
+	}
 
 	/*
 	 * Save nodes for contextualization: this will be used to "clone"
···
 	case MPOL_LOCAL:
 		break;
 	case MPOL_PREFERRED:
-		node_set(pol->v.preferred_node, nodes);
-		break;
 	case MPOL_BIND:
 	case MPOL_INTERLEAVE:
-		nodes = pol->v.nodes;
+		nodes = pol->nodes;
 		break;
 	default:
 		WARN_ON_ONCE(1);