xref: /freebsd/sys/vm/vm_radix.c (revision d6b92ffa)
1 /*
2  * Copyright (c) 2013 EMC Corp.
3  * Copyright (c) 2011 Jeffrey Roberson <jeff@freebsd.org>
4  * Copyright (c) 2008 Mayur Shardul <mayur.shardul@gmail.com>
5  * All rights reserved.
6  *
7  * Redistribution and use in source and binary forms, with or without
8  * modification, are permitted provided that the following conditions
9  * are met:
10  * 1. Redistributions of source code must retain the above copyright
11  *    notice, this list of conditions and the following disclaimer.
12  * 2. Redistributions in binary form must reproduce the above copyright
13  *    notice, this list of conditions and the following disclaimer in the
14  *    documentation and/or other materials provided with the distribution.
15  *
16  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
17  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
20  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
26  * SUCH DAMAGE.
27  *
28  */
29 
30 /*
31  * Path-compressed radix trie implementation.
32  * The following code is not generalized into a general purpose library
33  * because there are way too many parameters embedded that should really
34  * be decided by the library consumers.  At the same time, consumers
35  * of this code must achieve highest possible performance.
36  *
37  * The implementation takes into account the following rationale:
38  * - Size of the nodes should be as small as possible but still big enough
39  *   to avoid a large maximum depth for the trie.  This is a balance
40  *   between the necessity to not wire too much physical memory for the nodes
41  *   and the necessity to avoid too much cache pollution during the trie
42  *   operations.
43  * - There is not a huge bias toward the number of lookup operations over
44  *   the number of insert and remove operations.  This basically implies
45  *   that optimizations supposedly helping one operation but hurting the
46  *   other might be carefully evaluated.
47  * - On average not many nodes are expected to be fully populated, hence
48  *   level compression may just complicate things.
49  */
50 
51 #include <sys/cdefs.h>
52 __FBSDID("$FreeBSD$");
53 
54 #include "opt_ddb.h"
55 
56 #include <sys/param.h>
57 #include <sys/systm.h>
58 #include <sys/kernel.h>
59 #include <sys/vmmeter.h>
60 
61 #include <vm/uma.h>
62 #include <vm/vm.h>
63 #include <vm/vm_param.h>
64 #include <vm/vm_page.h>
65 #include <vm/vm_radix.h>
66 
67 #ifdef DDB
68 #include <ddb/ddb.h>
69 #endif
70 
71 /*
72  * These widths should allow the pointers to a node's children to fit within
73  * a single cache line.  The extra levels from a narrow width should not be
74  * a problem thanks to path compression.
75  */
76 #ifdef __LP64__
77 #define	VM_RADIX_WIDTH	4
78 #else
79 #define	VM_RADIX_WIDTH	3
80 #endif
81 
82 #define	VM_RADIX_COUNT	(1 << VM_RADIX_WIDTH)
83 #define	VM_RADIX_MASK	(VM_RADIX_COUNT - 1)
84 #define	VM_RADIX_LIMIT							\
85 	(howmany(sizeof(vm_pindex_t) * NBBY, VM_RADIX_WIDTH) - 1)
86 
87 /* Flag bits stored in node pointers. */
88 #define	VM_RADIX_ISLEAF	0x1
89 #define	VM_RADIX_FLAGS	0x1
90 #define	VM_RADIX_PAD	VM_RADIX_FLAGS
91 
92 /* Returns one unit associated with specified level. */
93 #define	VM_RADIX_UNITLEVEL(lev)						\
94 	((vm_pindex_t)1 << ((lev) * VM_RADIX_WIDTH))
95 
96 struct vm_radix_node {
97 	vm_pindex_t	 rn_owner;			/* Owner of record. */
98 	uint16_t	 rn_count;			/* Valid children. */
99 	uint16_t	 rn_clev;			/* Current level. */
100 	void		*rn_child[VM_RADIX_COUNT];	/* Child nodes. */
101 };
102 
103 static uma_zone_t vm_radix_node_zone;
104 
105 /*
106  * Allocate a radix node.
107  */
108 static __inline struct vm_radix_node *
109 vm_radix_node_get(vm_pindex_t owner, uint16_t count, uint16_t clevel)
110 {
111 	struct vm_radix_node *rnode;
112 
113 	rnode = uma_zalloc(vm_radix_node_zone, M_NOWAIT | M_ZERO);
114 	if (rnode == NULL)
115 		return (NULL);
116 	rnode->rn_owner = owner;
117 	rnode->rn_count = count;
118 	rnode->rn_clev = clevel;
119 	return (rnode);
120 }
121 
122 /*
123  * Free radix node.
124  */
125 static __inline void
126 vm_radix_node_put(struct vm_radix_node *rnode)
127 {
128 
129 	uma_zfree(vm_radix_node_zone, rnode);
130 }
131 
132 /*
133  * Return the position in the array for a given level.
134  */
135 static __inline int
136 vm_radix_slot(vm_pindex_t index, uint16_t level)
137 {
138 
139 	return ((index >> (level * VM_RADIX_WIDTH)) & VM_RADIX_MASK);
140 }
141 
142 /* Trims the key after the specified level. */
143 static __inline vm_pindex_t
144 vm_radix_trimkey(vm_pindex_t index, uint16_t level)
145 {
146 	vm_pindex_t ret;
147 
148 	ret = index;
149 	if (level > 0) {
150 		ret >>= level * VM_RADIX_WIDTH;
151 		ret <<= level * VM_RADIX_WIDTH;
152 	}
153 	return (ret);
154 }
155 
156 /*
157  * Get the root node for a radix tree.
158  */
159 static __inline struct vm_radix_node *
160 vm_radix_getroot(struct vm_radix *rtree)
161 {
162 
163 	return ((struct vm_radix_node *)rtree->rt_root);
164 }
165 
166 /*
167  * Set the root node for a radix tree.
168  */
169 static __inline void
170 vm_radix_setroot(struct vm_radix *rtree, struct vm_radix_node *rnode)
171 {
172 
173 	rtree->rt_root = (uintptr_t)rnode;
174 }
175 
176 /*
177  * Returns TRUE if the specified radix node is a leaf and FALSE otherwise.
178  */
179 static __inline boolean_t
180 vm_radix_isleaf(struct vm_radix_node *rnode)
181 {
182 
183 	return (((uintptr_t)rnode & VM_RADIX_ISLEAF) != 0);
184 }
185 
186 /*
187  * Returns the associated page extracted from rnode.
188  */
189 static __inline vm_page_t
190 vm_radix_topage(struct vm_radix_node *rnode)
191 {
192 
193 	return ((vm_page_t)((uintptr_t)rnode & ~VM_RADIX_FLAGS));
194 }
195 
196 /*
197  * Adds the page as a child of the provided node.
198  */
199 static __inline void
200 vm_radix_addpage(struct vm_radix_node *rnode, vm_pindex_t index, uint16_t clev,
201     vm_page_t page)
202 {
203 	int slot;
204 
205 	slot = vm_radix_slot(index, clev);
206 	rnode->rn_child[slot] = (void *)((uintptr_t)page | VM_RADIX_ISLEAF);
207 }
208 
209 /*
210  * Returns the slot where two keys differ.
211  * It cannot accept 2 equal keys.
212  */
213 static __inline uint16_t
214 vm_radix_keydiff(vm_pindex_t index1, vm_pindex_t index2)
215 {
216 	uint16_t clev;
217 
218 	KASSERT(index1 != index2, ("%s: passing the same key value %jx",
219 	    __func__, (uintmax_t)index1));
220 
221 	index1 ^= index2;
222 	for (clev = VM_RADIX_LIMIT;; clev--)
223 		if (vm_radix_slot(index1, clev) != 0)
224 			return (clev);
225 }
226 
227 /*
228  * Returns TRUE if it can be determined that key does not belong to the
229  * specified rnode.  Otherwise, returns FALSE.
230  */
231 static __inline boolean_t
232 vm_radix_keybarr(struct vm_radix_node *rnode, vm_pindex_t idx)
233 {
234 
235 	if (rnode->rn_clev < VM_RADIX_LIMIT) {
236 		idx = vm_radix_trimkey(idx, rnode->rn_clev + 1);
237 		return (idx != rnode->rn_owner);
238 	}
239 	return (FALSE);
240 }
241 
242 /*
243  * Internal helper for vm_radix_reclaim_allnodes().
244  * This function is recursive.
245  */
246 static void
247 vm_radix_reclaim_allnodes_int(struct vm_radix_node *rnode)
248 {
249 	int slot;
250 
251 	KASSERT(rnode->rn_count <= VM_RADIX_COUNT,
252 	    ("vm_radix_reclaim_allnodes_int: bad count in rnode %p", rnode));
253 	for (slot = 0; rnode->rn_count != 0; slot++) {
254 		if (rnode->rn_child[slot] == NULL)
255 			continue;
256 		if (!vm_radix_isleaf(rnode->rn_child[slot]))
257 			vm_radix_reclaim_allnodes_int(rnode->rn_child[slot]);
258 		rnode->rn_child[slot] = NULL;
259 		rnode->rn_count--;
260 	}
261 	vm_radix_node_put(rnode);
262 }
263 
264 #ifdef INVARIANTS
265 /*
266  * Radix node zone destructor.
267  */
268 static void
269 vm_radix_node_zone_dtor(void *mem, int size __unused, void *arg __unused)
270 {
271 	struct vm_radix_node *rnode;
272 	int slot;
273 
274 	rnode = mem;
275 	KASSERT(rnode->rn_count == 0,
276 	    ("vm_radix_node_put: rnode %p has %d children", rnode,
277 	    rnode->rn_count));
278 	for (slot = 0; slot < VM_RADIX_COUNT; slot++)
279 		KASSERT(rnode->rn_child[slot] == NULL,
280 		    ("vm_radix_node_put: rnode %p has a child", rnode));
281 }
282 #endif
283 
284 #ifndef UMA_MD_SMALL_ALLOC
285 /*
286  * Reserve the KVA necessary to satisfy the node allocation.
287  * This is mandatory in architectures not supporting direct
288  * mapping as they will need otherwise to carve into the kernel maps for
289  * every node allocation, resulting into deadlocks for consumers already
290  * working with kernel maps.
291  */
292 static void
293 vm_radix_reserve_kva(void *arg __unused)
294 {
295 
296 	/*
297 	 * Calculate the number of reserved nodes, discounting the pages that
298 	 * are needed to store them.
299 	 */
300 	if (!uma_zone_reserve_kva(vm_radix_node_zone,
301 	    ((vm_paddr_t)vm_cnt.v_page_count * PAGE_SIZE) / (PAGE_SIZE +
302 	    sizeof(struct vm_radix_node))))
303 		panic("%s: unable to reserve KVA", __func__);
304 }
305 SYSINIT(vm_radix_reserve_kva, SI_SUB_KMEM, SI_ORDER_THIRD,
306     vm_radix_reserve_kva, NULL);
307 #endif
308 
309 /*
310  * Initialize the UMA slab zone.
311  */
312 void
313 vm_radix_zinit(void)
314 {
315 
316 	vm_radix_node_zone = uma_zcreate("RADIX NODE",
317 	    sizeof(struct vm_radix_node), NULL,
318 #ifdef INVARIANTS
319 	    vm_radix_node_zone_dtor,
320 #else
321 	    NULL,
322 #endif
323 	    NULL, NULL, VM_RADIX_PAD, UMA_ZONE_VM);
324 }
325 
326 /*
327  * Inserts the key-value pair into the trie.
328  * Panics if the key already exists.
329  */
330 int
331 vm_radix_insert(struct vm_radix *rtree, vm_page_t page)
332 {
333 	vm_pindex_t index, newind;
334 	void **parentp;
335 	struct vm_radix_node *rnode, *tmp;
336 	vm_page_t m;
337 	int slot;
338 	uint16_t clev;
339 
340 	index = page->pindex;
341 
342 	/*
343 	 * The owner of record for root is not really important because it
344 	 * will never be used.
345 	 */
346 	rnode = vm_radix_getroot(rtree);
347 	if (rnode == NULL) {
348 		rtree->rt_root = (uintptr_t)page | VM_RADIX_ISLEAF;
349 		return (0);
350 	}
351 	parentp = (void **)&rtree->rt_root;
352 	for (;;) {
353 		if (vm_radix_isleaf(rnode)) {
354 			m = vm_radix_topage(rnode);
355 			if (m->pindex == index)
356 				panic("%s: key %jx is already present",
357 				    __func__, (uintmax_t)index);
358 			clev = vm_radix_keydiff(m->pindex, index);
359 			tmp = vm_radix_node_get(vm_radix_trimkey(index,
360 			    clev + 1), 2, clev);
361 			if (tmp == NULL)
362 				return (ENOMEM);
363 			*parentp = tmp;
364 			vm_radix_addpage(tmp, index, clev, page);
365 			vm_radix_addpage(tmp, m->pindex, clev, m);
366 			return (0);
367 		} else if (vm_radix_keybarr(rnode, index))
368 			break;
369 		slot = vm_radix_slot(index, rnode->rn_clev);
370 		if (rnode->rn_child[slot] == NULL) {
371 			rnode->rn_count++;
372 			vm_radix_addpage(rnode, index, rnode->rn_clev, page);
373 			return (0);
374 		}
375 		parentp = &rnode->rn_child[slot];
376 		rnode = rnode->rn_child[slot];
377 	}
378 
379 	/*
380 	 * A new node is needed because the right insertion level is reached.
381 	 * Setup the new intermediate node and add the 2 children: the
382 	 * new object and the older edge.
383 	 */
384 	newind = rnode->rn_owner;
385 	clev = vm_radix_keydiff(newind, index);
386 	tmp = vm_radix_node_get(vm_radix_trimkey(index, clev + 1), 2, clev);
387 	if (tmp == NULL)
388 		return (ENOMEM);
389 	*parentp = tmp;
390 	vm_radix_addpage(tmp, index, clev, page);
391 	slot = vm_radix_slot(newind, clev);
392 	tmp->rn_child[slot] = rnode;
393 	return (0);
394 }
395 
396 /*
397  * Returns TRUE if the specified radix tree contains a single leaf and FALSE
398  * otherwise.
399  */
400 boolean_t
401 vm_radix_is_singleton(struct vm_radix *rtree)
402 {
403 	struct vm_radix_node *rnode;
404 
405 	rnode = vm_radix_getroot(rtree);
406 	if (rnode == NULL)
407 		return (FALSE);
408 	return (vm_radix_isleaf(rnode));
409 }
410 
411 /*
412  * Returns the value stored at the index.  If the index is not present,
413  * NULL is returned.
414  */
415 vm_page_t
416 vm_radix_lookup(struct vm_radix *rtree, vm_pindex_t index)
417 {
418 	struct vm_radix_node *rnode;
419 	vm_page_t m;
420 	int slot;
421 
422 	rnode = vm_radix_getroot(rtree);
423 	while (rnode != NULL) {
424 		if (vm_radix_isleaf(rnode)) {
425 			m = vm_radix_topage(rnode);
426 			if (m->pindex == index)
427 				return (m);
428 			else
429 				break;
430 		} else if (vm_radix_keybarr(rnode, index))
431 			break;
432 		slot = vm_radix_slot(index, rnode->rn_clev);
433 		rnode = rnode->rn_child[slot];
434 	}
435 	return (NULL);
436 }
437 
438 /*
439  * Look up the nearest entry at a position bigger than or equal to index.
440  */
441 vm_page_t
442 vm_radix_lookup_ge(struct vm_radix *rtree, vm_pindex_t index)
443 {
444 	struct vm_radix_node *stack[VM_RADIX_LIMIT];
445 	vm_pindex_t inc;
446 	vm_page_t m;
447 	struct vm_radix_node *child, *rnode;
448 #ifdef INVARIANTS
449 	int loops = 0;
450 #endif
451 	int slot, tos;
452 
453 	rnode = vm_radix_getroot(rtree);
454 	if (rnode == NULL)
455 		return (NULL);
456 	else if (vm_radix_isleaf(rnode)) {
457 		m = vm_radix_topage(rnode);
458 		if (m->pindex >= index)
459 			return (m);
460 		else
461 			return (NULL);
462 	}
463 	tos = 0;
464 	for (;;) {
465 		/*
466 		 * If the keys differ before the current bisection node,
467 		 * then the search key might rollback to the earliest
468 		 * available bisection node or to the smallest key
469 		 * in the current node (if the owner is bigger than the
470 		 * search key).
471 		 */
472 		if (vm_radix_keybarr(rnode, index)) {
473 			if (index > rnode->rn_owner) {
474 ascend:
475 				KASSERT(++loops < 1000,
476 				    ("vm_radix_lookup_ge: too many loops"));
477 
478 				/*
479 				 * Pop nodes from the stack until either the
480 				 * stack is empty or a node that could have a
481 				 * matching descendant is found.
482 				 */
483 				do {
484 					if (tos == 0)
485 						return (NULL);
486 					rnode = stack[--tos];
487 				} while (vm_radix_slot(index,
488 				    rnode->rn_clev) == (VM_RADIX_COUNT - 1));
489 
490 				/*
491 				 * The following computation cannot overflow
492 				 * because index's slot at the current level
493 				 * is less than VM_RADIX_COUNT - 1.
494 				 */
495 				index = vm_radix_trimkey(index,
496 				    rnode->rn_clev);
497 				index += VM_RADIX_UNITLEVEL(rnode->rn_clev);
498 			} else
499 				index = rnode->rn_owner;
500 			KASSERT(!vm_radix_keybarr(rnode, index),
501 			    ("vm_radix_lookup_ge: keybarr failed"));
502 		}
503 		slot = vm_radix_slot(index, rnode->rn_clev);
504 		child = rnode->rn_child[slot];
505 		if (vm_radix_isleaf(child)) {
506 			m = vm_radix_topage(child);
507 			if (m->pindex >= index)
508 				return (m);
509 		} else if (child != NULL)
510 			goto descend;
511 
512 		/*
513 		 * Look for an available edge or page within the current
514 		 * bisection node.
515 		 */
516                 if (slot < (VM_RADIX_COUNT - 1)) {
517 			inc = VM_RADIX_UNITLEVEL(rnode->rn_clev);
518 			index = vm_radix_trimkey(index, rnode->rn_clev);
519 			do {
520 				index += inc;
521 				slot++;
522 				child = rnode->rn_child[slot];
523 				if (vm_radix_isleaf(child)) {
524 					m = vm_radix_topage(child);
525 					if (m->pindex >= index)
526 						return (m);
527 				} else if (child != NULL)
528 					goto descend;
529 			} while (slot < (VM_RADIX_COUNT - 1));
530 		}
531 		KASSERT(child == NULL || vm_radix_isleaf(child),
532 		    ("vm_radix_lookup_ge: child is radix node"));
533 
534 		/*
535 		 * If a page or edge bigger than the search slot is not found
536 		 * in the current node, ascend to the next higher-level node.
537 		 */
538 		goto ascend;
539 descend:
540 		KASSERT(rnode->rn_clev > 0,
541 		    ("vm_radix_lookup_ge: pushing leaf's parent"));
542 		KASSERT(tos < VM_RADIX_LIMIT,
543 		    ("vm_radix_lookup_ge: stack overflow"));
544 		stack[tos++] = rnode;
545 		rnode = child;
546 	}
547 }
548 
549 /*
550  * Look up the nearest entry at a position less than or equal to index.
551  */
552 vm_page_t
553 vm_radix_lookup_le(struct vm_radix *rtree, vm_pindex_t index)
554 {
555 	struct vm_radix_node *stack[VM_RADIX_LIMIT];
556 	vm_pindex_t inc;
557 	vm_page_t m;
558 	struct vm_radix_node *child, *rnode;
559 #ifdef INVARIANTS
560 	int loops = 0;
561 #endif
562 	int slot, tos;
563 
564 	rnode = vm_radix_getroot(rtree);
565 	if (rnode == NULL)
566 		return (NULL);
567 	else if (vm_radix_isleaf(rnode)) {
568 		m = vm_radix_topage(rnode);
569 		if (m->pindex <= index)
570 			return (m);
571 		else
572 			return (NULL);
573 	}
574 	tos = 0;
575 	for (;;) {
576 		/*
577 		 * If the keys differ before the current bisection node,
578 		 * then the search key might rollback to the earliest
579 		 * available bisection node or to the largest key
580 		 * in the current node (if the owner is smaller than the
581 		 * search key).
582 		 */
583 		if (vm_radix_keybarr(rnode, index)) {
584 			if (index > rnode->rn_owner) {
585 				index = rnode->rn_owner + VM_RADIX_COUNT *
586 				    VM_RADIX_UNITLEVEL(rnode->rn_clev);
587 			} else {
588 ascend:
589 				KASSERT(++loops < 1000,
590 				    ("vm_radix_lookup_le: too many loops"));
591 
592 				/*
593 				 * Pop nodes from the stack until either the
594 				 * stack is empty or a node that could have a
595 				 * matching descendant is found.
596 				 */
597 				do {
598 					if (tos == 0)
599 						return (NULL);
600 					rnode = stack[--tos];
601 				} while (vm_radix_slot(index,
602 				    rnode->rn_clev) == 0);
603 
604 				/*
605 				 * The following computation cannot overflow
606 				 * because index's slot at the current level
607 				 * is greater than 0.
608 				 */
609 				index = vm_radix_trimkey(index,
610 				    rnode->rn_clev);
611 			}
612 			index--;
613 			KASSERT(!vm_radix_keybarr(rnode, index),
614 			    ("vm_radix_lookup_le: keybarr failed"));
615 		}
616 		slot = vm_radix_slot(index, rnode->rn_clev);
617 		child = rnode->rn_child[slot];
618 		if (vm_radix_isleaf(child)) {
619 			m = vm_radix_topage(child);
620 			if (m->pindex <= index)
621 				return (m);
622 		} else if (child != NULL)
623 			goto descend;
624 
625 		/*
626 		 * Look for an available edge or page within the current
627 		 * bisection node.
628 		 */
629 		if (slot > 0) {
630 			inc = VM_RADIX_UNITLEVEL(rnode->rn_clev);
631 			index |= inc - 1;
632 			do {
633 				index -= inc;
634 				slot--;
635 				child = rnode->rn_child[slot];
636 				if (vm_radix_isleaf(child)) {
637 					m = vm_radix_topage(child);
638 					if (m->pindex <= index)
639 						return (m);
640 				} else if (child != NULL)
641 					goto descend;
642 			} while (slot > 0);
643 		}
644 		KASSERT(child == NULL || vm_radix_isleaf(child),
645 		    ("vm_radix_lookup_le: child is radix node"));
646 
647 		/*
648 		 * If a page or edge smaller than the search slot is not found
649 		 * in the current node, ascend to the next higher-level node.
650 		 */
651 		goto ascend;
652 descend:
653 		KASSERT(rnode->rn_clev > 0,
654 		    ("vm_radix_lookup_le: pushing leaf's parent"));
655 		KASSERT(tos < VM_RADIX_LIMIT,
656 		    ("vm_radix_lookup_le: stack overflow"));
657 		stack[tos++] = rnode;
658 		rnode = child;
659 	}
660 }
661 
662 /*
663  * Remove the specified index from the trie, and return the value stored at
664  * that index.  If the index is not present, return NULL.
665  */
666 vm_page_t
667 vm_radix_remove(struct vm_radix *rtree, vm_pindex_t index)
668 {
669 	struct vm_radix_node *rnode, *parent;
670 	vm_page_t m;
671 	int i, slot;
672 
673 	rnode = vm_radix_getroot(rtree);
674 	if (vm_radix_isleaf(rnode)) {
675 		m = vm_radix_topage(rnode);
676 		if (m->pindex != index)
677 			return (NULL);
678 		vm_radix_setroot(rtree, NULL);
679 		return (m);
680 	}
681 	parent = NULL;
682 	for (;;) {
683 		if (rnode == NULL)
684 			return (NULL);
685 		slot = vm_radix_slot(index, rnode->rn_clev);
686 		if (vm_radix_isleaf(rnode->rn_child[slot])) {
687 			m = vm_radix_topage(rnode->rn_child[slot]);
688 			if (m->pindex != index)
689 				return (NULL);
690 			rnode->rn_child[slot] = NULL;
691 			rnode->rn_count--;
692 			if (rnode->rn_count > 1)
693 				return (m);
694 			for (i = 0; i < VM_RADIX_COUNT; i++)
695 				if (rnode->rn_child[i] != NULL)
696 					break;
697 			KASSERT(i != VM_RADIX_COUNT,
698 			    ("%s: invalid node configuration", __func__));
699 			if (parent == NULL)
700 				vm_radix_setroot(rtree, rnode->rn_child[i]);
701 			else {
702 				slot = vm_radix_slot(index, parent->rn_clev);
703 				KASSERT(parent->rn_child[slot] == rnode,
704 				    ("%s: invalid child value", __func__));
705 				parent->rn_child[slot] = rnode->rn_child[i];
706 			}
707 			rnode->rn_count--;
708 			rnode->rn_child[i] = NULL;
709 			vm_radix_node_put(rnode);
710 			return (m);
711 		}
712 		parent = rnode;
713 		rnode = rnode->rn_child[slot];
714 	}
715 }
716 
717 /*
718  * Remove and free all the nodes from the radix tree.
719  * This function is recursive but there is a tight control on it as the
720  * maximum depth of the tree is fixed.
721  */
722 void
723 vm_radix_reclaim_allnodes(struct vm_radix *rtree)
724 {
725 	struct vm_radix_node *root;
726 
727 	root = vm_radix_getroot(rtree);
728 	if (root == NULL)
729 		return;
730 	vm_radix_setroot(rtree, NULL);
731 	if (!vm_radix_isleaf(root))
732 		vm_radix_reclaim_allnodes_int(root);
733 }
734 
735 /*
736  * Replace an existing page in the trie with another one.
737  * Panics if there is not an old page in the trie at the new page's index.
738  */
739 vm_page_t
740 vm_radix_replace(struct vm_radix *rtree, vm_page_t newpage)
741 {
742 	struct vm_radix_node *rnode;
743 	vm_page_t m;
744 	vm_pindex_t index;
745 	int slot;
746 
747 	index = newpage->pindex;
748 	rnode = vm_radix_getroot(rtree);
749 	if (rnode == NULL)
750 		panic("%s: replacing page on an empty trie", __func__);
751 	if (vm_radix_isleaf(rnode)) {
752 		m = vm_radix_topage(rnode);
753 		if (m->pindex != index)
754 			panic("%s: original replacing root key not found",
755 			    __func__);
756 		rtree->rt_root = (uintptr_t)newpage | VM_RADIX_ISLEAF;
757 		return (m);
758 	}
759 	for (;;) {
760 		slot = vm_radix_slot(index, rnode->rn_clev);
761 		if (vm_radix_isleaf(rnode->rn_child[slot])) {
762 			m = vm_radix_topage(rnode->rn_child[slot]);
763 			if (m->pindex == index) {
764 				rnode->rn_child[slot] =
765 				    (void *)((uintptr_t)newpage |
766 				    VM_RADIX_ISLEAF);
767 				return (m);
768 			} else
769 				break;
770 		} else if (rnode->rn_child[slot] == NULL ||
771 		    vm_radix_keybarr(rnode->rn_child[slot], index))
772 			break;
773 		rnode = rnode->rn_child[slot];
774 	}
775 	panic("%s: original replacing page not found", __func__);
776 }
777 
778 #ifdef DDB
779 /*
780  * Show details about the given radix node.
781  */
782 DB_SHOW_COMMAND(radixnode, db_show_radixnode)
783 {
784 	struct vm_radix_node *rnode;
785 	int i;
786 
787         if (!have_addr)
788                 return;
789 	rnode = (struct vm_radix_node *)addr;
790 	db_printf("radixnode %p, owner %jx, children count %u, level %u:\n",
791 	    (void *)rnode, (uintmax_t)rnode->rn_owner, rnode->rn_count,
792 	    rnode->rn_clev);
793 	for (i = 0; i < VM_RADIX_COUNT; i++)
794 		if (rnode->rn_child[i] != NULL)
795 			db_printf("slot: %d, val: %p, page: %p, clev: %d\n",
796 			    i, (void *)rnode->rn_child[i],
797 			    vm_radix_isleaf(rnode->rn_child[i]) ?
798 			    vm_radix_topage(rnode->rn_child[i]) : NULL,
799 			    rnode->rn_clev);
800 }
801 #endif /* DDB */
802