summaryrefslogtreecommitdiff
path: root/sys/kern/kern_malloc.c
blob: eb3f62fb99ab7b03aa6e0723a62a3ed836fa4f24 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
/*	$OpenBSD: kern_malloc.c,v 1.128 2015/03/14 03:38:50 jsg Exp $	*/
/*	$NetBSD: kern_malloc.c,v 1.15.4.2 1996/06/13 17:10:56 cgd Exp $	*/

/*
 * Copyright (c) 1987, 1991, 1993
 *	The Regents of the University of California.  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	@(#)kern_malloc.c	8.3 (Berkeley) 1/4/94
 */

#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <sys/stdint.h>
#include <sys/systm.h>
#include <sys/sysctl.h>
#include <sys/time.h>
#include <sys/rwlock.h>

#include <uvm/uvm_extern.h>

static
#ifndef SMALL_KERNEL
__inline__
#endif
long BUCKETINDX(size_t sz)
{
	long b, d;

	/* note that this relies upon MINALLOCSIZE being 1 << MINBUCKET */
	b = 7 + MINBUCKET; d = 4;
	while (d != 0) {
		if (sz <= (1 << b))
			b -= d;
		else
			b += d;
		d >>= 1;
	}
	if (sz <= (1 << b))
		b += 0;
	else
		b += 1;
	return b;
}

static struct vm_map kmem_map_store;
struct vm_map *kmem_map = NULL;

/*
 * Default number of pages in kmem_map.  We attempt to calculate this
 * at run-time, but allow it to be either patched or set in the kernel
 * config file.
 */
#ifndef NKMEMPAGES
#define	NKMEMPAGES	0
#endif
u_int	nkmempages = NKMEMPAGES;

/*
 * Defaults for lower- and upper-bounds for the kmem_map page count.
 * Can be overridden by kernel config options.
 */
#ifndef	NKMEMPAGES_MIN
#define	NKMEMPAGES_MIN	0
#endif
u_int	nkmempages_min = 0;

#ifndef NKMEMPAGES_MAX
#define	NKMEMPAGES_MAX	NKMEMPAGES_MAX_DEFAULT
#endif
u_int	nkmempages_max = 0;

struct kmembuckets bucket[MINBUCKET + 16];
#ifdef KMEMSTATS
struct kmemstats kmemstats[M_LAST];
#endif
struct kmemusage *kmemusage;
char *kmembase, *kmemlimit;
char buckstring[16 * sizeof("123456,")];
int buckstring_init = 0;
#if defined(KMEMSTATS) || defined(DIAGNOSTIC) || defined(FFS_SOFTUPDATES)
char *memname[] = INITKMEMNAMES;
char *memall = NULL;
struct rwlock sysctl_kmemlock = RWLOCK_INITIALIZER("sysctlklk");
#endif

/*
 * Normally the freelist structure is used only to hold the list pointer
 * for free objects.  However, when running with diagnostics, the first
 * 8 bytes of the structure is unused except for diagnostic information,
 * and the free list pointer is at offset 8 in the structure.  Since the
 * first 8 bytes is the portion of the structure most often modified, this
 * helps to detect memory reuse problems and avoid free list corruption.
 */
struct kmem_freelist {
	int32_t	kf_spare0;
	int16_t	kf_type;
	int16_t	kf_spare1;
	XSIMPLEQ_ENTRY(kmem_freelist) kf_flist;
};

#ifdef DIAGNOSTIC
/*
 * This structure provides a set of masks to catch unaligned frees.
 */
const long addrmask[] = { 0,
	0x00000001, 0x00000003, 0x00000007, 0x0000000f,
	0x0000001f, 0x0000003f, 0x0000007f, 0x000000ff,
	0x000001ff, 0x000003ff, 0x000007ff, 0x00000fff,
	0x00001fff, 0x00003fff, 0x00007fff, 0x0000ffff,
};

#endif /* DIAGNOSTIC */

#ifndef SMALL_KERNEL
struct timeval malloc_errintvl = { 5, 0 };
struct timeval malloc_lasterr;
#endif

/*
 * Allocate a block of memory
 */
void *
malloc(size_t size, int type, int flags)
{
	struct kmembuckets *kbp;
	struct kmemusage *kup;
	struct kmem_freelist *freep;
	long indx, npg, allocsize;
	int s;
	caddr_t va, cp;
#ifdef DIAGNOSTIC
	int freshalloc;
	char *savedtype;
#endif
#ifdef KMEMSTATS
	struct kmemstats *ksp = &kmemstats[type];

	if (((unsigned long)type) <= 1 || ((unsigned long)type) >= M_LAST)
		panic("malloc: bogus type %d", type);
#endif

	KASSERT(flags & (M_WAITOK | M_NOWAIT));

	if ((flags & M_NOWAIT) == 0) {
		extern int pool_debug;
#ifdef DIAGNOSTIC
		assertwaitok();
		if (pool_debug == 2)
			yield();
#endif
		if (!cold && pool_debug) {
			KERNEL_UNLOCK();
			KERNEL_LOCK();
		}
	}

#ifdef MALLOC_DEBUG
	if (debug_malloc(size, type, flags, (void **)&va)) {
		if ((flags & M_ZERO) && va != NULL)
			memset(va, 0, size);
		return (va);
	}
#endif

	if (size > 65535 * PAGE_SIZE) {
		if (flags & M_CANFAIL) {
#ifndef SMALL_KERNEL
			if (ratecheck(&malloc_lasterr, &malloc_errintvl))
				printf("malloc(): allocation too large, "
				    "type = %d, size = %lu\n", type, size);
#endif
			return (NULL);
		} else
			panic("malloc: allocation too large, "
			    "type = %d, size = %lu\n", type, size);
	}

	indx = BUCKETINDX(size);
	kbp = &bucket[indx];
	s = splvm();
#ifdef KMEMSTATS
	while (ksp->ks_memuse >= ksp->ks_limit) {
		if (flags & M_NOWAIT) {
			splx(s);
			return (NULL);
		}
		if (ksp->ks_limblocks < 65535)
			ksp->ks_limblocks++;
		tsleep(ksp, PSWP+2, memname[type], 0);
	}
	ksp->ks_size |= 1 << indx;
#endif
	if (size > MAXALLOCSAVE)
		allocsize = round_page(size);
	else
		allocsize = 1 << indx;
	if (XSIMPLEQ_FIRST(&kbp->kb_freelist) == NULL) {
		npg = atop(round_page(allocsize));
		va = (caddr_t)uvm_km_kmemalloc_pla(kmem_map, NULL,
		    (vsize_t)ptoa(npg), 0,
		    ((flags & M_NOWAIT) ? UVM_KMF_NOWAIT : 0) |
		    ((flags & M_CANFAIL) ? UVM_KMF_CANFAIL : 0),
		    no_constraint.ucr_low, no_constraint.ucr_high,
		    0, 0, 0);
		if (va == NULL) {
			/*
			 * Kmem_malloc() can return NULL, even if it can
			 * wait, if there is no map space available, because
			 * it can't fix that problem.  Neither can we,
			 * right now.  (We should release pages which
			 * are completely free and which are in buckets
			 * with too many free elements.)
			 */
			if ((flags & (M_NOWAIT|M_CANFAIL)) == 0)
				panic("malloc: out of space in kmem_map");
			splx(s);
			return (NULL);
		}
#ifdef KMEMSTATS
		kbp->kb_total += kbp->kb_elmpercl;
#endif
		kup = btokup(va);
		kup->ku_indx = indx;
#ifdef DIAGNOSTIC
		freshalloc = 1;
#endif
		if (allocsize > MAXALLOCSAVE) {
			kup->ku_pagecnt = npg;
#ifdef KMEMSTATS
			ksp->ks_memuse += allocsize;
#endif
			goto out;
		}
#ifdef KMEMSTATS
		kup->ku_freecnt = kbp->kb_elmpercl;
		kbp->kb_totalfree += kbp->kb_elmpercl;
#endif
		cp = va + (npg * PAGE_SIZE) - allocsize;
		for (;;) {
			freep = (struct kmem_freelist *)cp;
#ifdef DIAGNOSTIC
			/*
			 * Copy in known text to detect modification
			 * after freeing.
			 */
			poison_mem(cp, allocsize);
			freep->kf_type = M_FREE;
#endif /* DIAGNOSTIC */
			XSIMPLEQ_INSERT_HEAD(&kbp->kb_freelist, freep, kf_flist);
			if (cp <= va)
				break;
			cp -= allocsize;
		}
	} else {
#ifdef DIAGNOSTIC
		freshalloc = 0;
#endif
	}
	freep = XSIMPLEQ_FIRST(&kbp->kb_freelist);
	XSIMPLEQ_REMOVE_HEAD(&kbp->kb_freelist, kf_flist);
	va = (caddr_t)freep;
#ifdef DIAGNOSTIC
	savedtype = (unsigned)freep->kf_type < M_LAST ?
		memname[freep->kf_type] : "???";
	if (freshalloc == 0 && XSIMPLEQ_FIRST(&kbp->kb_freelist)) {
		int rv;
		vaddr_t addr = (vaddr_t)XSIMPLEQ_FIRST(&kbp->kb_freelist);

		vm_map_lock(kmem_map);
		rv = uvm_map_checkprot(kmem_map, addr,
		    addr + sizeof(struct kmem_freelist), PROT_WRITE);
		vm_map_unlock(kmem_map);

		if (!rv)  {
			printf("%s %zd of object %p size 0x%lx %s %s"
			    " (invalid addr %p)\n",
			    "Data modified on freelist: word", 
			    (int32_t *)&addr - (int32_t *)kbp, va, size,
			    "previous type", savedtype, (void *)addr);
		}
	}

	/* Fill the fields that we've used with poison */
	poison_mem(freep, sizeof(*freep));

	/* and check that the data hasn't been modified. */
	if (freshalloc == 0) {
		size_t pidx;
		uint32_t pval;
		if (poison_check(va, allocsize, &pidx, &pval)) {
			panic("%s %zd of object %p size 0x%lx %s %s"
			    " (0x%x != 0x%x)\n",
			    "Data modified on freelist: word",
			    pidx, va, size, "previous type",
			    savedtype, ((int32_t*)va)[pidx], pval);
		}
	}

	freep->kf_spare0 = 0;
#endif /* DIAGNOSTIC */
#ifdef KMEMSTATS
	kup = btokup(va);
	if (kup->ku_indx != indx)
		panic("malloc: wrong bucket");
	if (kup->ku_freecnt == 0)
		panic("malloc: lost data");
	kup->ku_freecnt--;
	kbp->kb_totalfree--;
	ksp->ks_memuse += 1 << indx;
out:
	kbp->kb_calls++;
	ksp->ks_inuse++;
	ksp->ks_calls++;
	if (ksp->ks_memuse > ksp->ks_maxused)
		ksp->ks_maxused = ksp->ks_memuse;
#else
out:
#endif
	splx(s);

	if ((flags & M_ZERO) && va != NULL)
		memset(va, 0, size);
	return (va);
}

/*
 * Free a block of memory allocated by malloc.
 */
void
free(void *addr, int type, size_t freedsize)
{
	struct kmembuckets *kbp;
	struct kmemusage *kup;
	struct kmem_freelist *freep;
	long size;
	int s;
#ifdef DIAGNOSTIC
	long alloc;
#endif
#ifdef KMEMSTATS
	struct kmemstats *ksp = &kmemstats[type];
#endif

	if (addr == NULL)
		return;

#ifdef MALLOC_DEBUG
	if (debug_free(addr, type))
		return;
#endif

#ifdef DIAGNOSTIC
	if (addr < (void *)kmembase || addr >= (void *)kmemlimit)
		panic("free: non-malloced addr %p type %s", addr,
		    memname[type]);
#endif

	kup = btokup(addr);
	size = 1 << kup->ku_indx;
	kbp = &bucket[kup->ku_indx];
	if (size > MAXALLOCSAVE)
		size = kup->ku_pagecnt << PAGE_SHIFT;
	s = splvm();
#ifdef DIAGNOSTIC
	if (freedsize != 0 && freedsize > size)
		panic("free: size too large %zu > %ld (%p) type %s",
		    freedsize, size, addr, memname[type]);
	if (freedsize != 0 && size > MINALLOCSIZE && freedsize < size / 2)
		panic("free: size too small %zu < %ld / 2 (%p) type %s",
		    freedsize, size, addr, memname[type]);
	/*
	 * Check for returns of data that do not point to the
	 * beginning of the allocation.
	 */
	if (size > PAGE_SIZE)
		alloc = addrmask[BUCKETINDX(PAGE_SIZE)];
	else
		alloc = addrmask[kup->ku_indx];
	if (((u_long)addr & alloc) != 0)
		panic("free: unaligned addr %p, size %ld, type %s, mask %ld",
			addr, size, memname[type], alloc);
#endif /* DIAGNOSTIC */
	if (size > MAXALLOCSAVE) {
		uvm_km_free(kmem_map, (vaddr_t)addr, ptoa(kup->ku_pagecnt));
#ifdef KMEMSTATS
		ksp->ks_memuse -= size;
		kup->ku_indx = 0;
		kup->ku_pagecnt = 0;
		if (ksp->ks_memuse + size >= ksp->ks_limit &&
		    ksp->ks_memuse < ksp->ks_limit)
			wakeup(ksp);
		ksp->ks_inuse--;
		kbp->kb_total -= 1;
#endif
		splx(s);
		return;
	}
	freep = (struct kmem_freelist *)addr;
#ifdef DIAGNOSTIC
	/*
	 * Check for multiple frees. Use a quick check to see if
	 * it looks free before laboriously searching the freelist.
	 */
	if (freep->kf_spare0 == poison_value(freep)) {
		struct kmem_freelist *fp;
		XSIMPLEQ_FOREACH(fp, &kbp->kb_freelist, kf_flist) {
			if (addr != fp)
				continue;
			printf("multiply freed item %p\n", addr);
			panic("free: duplicated free");
		}
	}
	/*
	 * Copy in known text to detect modification after freeing
	 * and to make it look free. Also, save the type being freed
	 * so we can list likely culprit if modification is detected
	 * when the object is reallocated.
	 */
	poison_mem(addr, size);
	freep->kf_spare0 = poison_value(freep);

	freep->kf_type = type;
#endif /* DIAGNOSTIC */
#ifdef KMEMSTATS
	kup->ku_freecnt++;
	if (kup->ku_freecnt >= kbp->kb_elmpercl) {
		if (kup->ku_freecnt > kbp->kb_elmpercl)
			panic("free: multiple frees");
		else if (kbp->kb_totalfree > kbp->kb_highwat)
			kbp->kb_couldfree++;
	}
	kbp->kb_totalfree++;
	ksp->ks_memuse -= size;
	if (ksp->ks_memuse + size >= ksp->ks_limit &&
	    ksp->ks_memuse < ksp->ks_limit)
		wakeup(ksp);
	ksp->ks_inuse--;
#endif
	XSIMPLEQ_INSERT_TAIL(&kbp->kb_freelist, freep, kf_flist);
	splx(s);
}

/*
 * Compute the number of pages that kmem_map will map, that is,
 * the size of the kernel malloc arena.
 */
void
kmeminit_nkmempages(void)
{
	u_int npages;

	if (nkmempages != 0) {
		/*
		 * It's already been set (by us being here before, or
		 * by patching or kernel config options), bail out now.
		 */
		return;
	}

	/*
	 * We can't initialize these variables at compilation time, since
	 * the page size may not be known (on sparc GENERIC kernels, for
	 * example). But we still want the MD code to be able to provide
	 * better values.
	 */
	if (nkmempages_min == 0)
		nkmempages_min = NKMEMPAGES_MIN;
	if (nkmempages_max == 0)
		nkmempages_max = NKMEMPAGES_MAX;

	/*
	 * We use the following (simple) formula:
	 *
	 *	- Starting point is physical memory / 4.
	 *
	 *	- Clamp it down to nkmempages_max.
	 *
	 *	- Round it up to nkmempages_min.
	 */
	npages = physmem / 4;

	if (npages > nkmempages_max)
		npages = nkmempages_max;

	if (npages < nkmempages_min)
		npages = nkmempages_min;

	nkmempages = npages;
}

/*
 * Initialize the kernel memory allocator
 */
void
kmeminit(void)
{
	vaddr_t base, limit;
	long indx;

#ifdef DIAGNOSTIC
	if (sizeof(struct kmem_freelist) > (1 << MINBUCKET))
		panic("kmeminit: minbucket too small/struct freelist too big");
#endif

	/*
	 * Compute the number of kmem_map pages, if we have not
	 * done so already.
	 */
	kmeminit_nkmempages();
	base = vm_map_min(kernel_map);
	kmem_map = uvm_km_suballoc(kernel_map, &base, &limit,
	    (vsize_t)nkmempages << PAGE_SHIFT,
#ifdef KVA_GUARDPAGES
	    VM_MAP_INTRSAFE | VM_MAP_GUARDPAGES,
#else
	    VM_MAP_INTRSAFE,
#endif
	    FALSE, &kmem_map_store);
	kmembase = (char *)base;
	kmemlimit = (char *)limit;
	kmemusage = (struct kmemusage *) uvm_km_zalloc(kernel_map,
		(vsize_t)(nkmempages * sizeof(struct kmemusage)));
	for (indx = 0; indx < MINBUCKET + 16; indx++) {
		XSIMPLEQ_INIT(&bucket[indx].kb_freelist);
	}
#ifdef KMEMSTATS
	for (indx = 0; indx < MINBUCKET + 16; indx++) {
		if (1 << indx >= PAGE_SIZE)
			bucket[indx].kb_elmpercl = 1;
		else
			bucket[indx].kb_elmpercl = PAGE_SIZE / (1 << indx);
		bucket[indx].kb_highwat = 5 * bucket[indx].kb_elmpercl;
	}
	for (indx = 0; indx < M_LAST; indx++)
		kmemstats[indx].ks_limit = nkmempages * PAGE_SIZE * 6 / 10;
#endif
#ifdef MALLOC_DEBUG
	debug_malloc_init();
#endif
}

/*
 * Return kernel malloc statistics information.
 */
int
sysctl_malloc(int *name, u_int namelen, void *oldp, size_t *oldlenp, void *newp,
    size_t newlen, struct proc *p)
{
	struct kmembuckets kb;
#if defined(KMEMSTATS) || defined(DIAGNOSTIC) || defined(FFS_SOFTUPDATES)
	int error;
#endif
	int i, siz;

	if (namelen != 2 && name[0] != KERN_MALLOC_BUCKETS &&
	    name[0] != KERN_MALLOC_KMEMNAMES)
		return (ENOTDIR);		/* overloaded */

	switch (name[0]) {
	case KERN_MALLOC_BUCKETS:
		/* Initialize the first time */
		if (buckstring_init == 0) {
			buckstring_init = 1;
			memset(buckstring, 0, sizeof(buckstring));
			for (siz = 0, i = MINBUCKET; i < MINBUCKET + 16; i++) {
				snprintf(buckstring + siz,
				    sizeof buckstring - siz,
				    "%d,", (u_int)(1<<i));
				siz += strlen(buckstring + siz);
			}
			/* Remove trailing comma */
			if (siz)
				buckstring[siz - 1] = '\0';
		}
		return (sysctl_rdstring(oldp, oldlenp, newp, buckstring));

	case KERN_MALLOC_BUCKET:
		memcpy(&kb, &bucket[BUCKETINDX(name[1])], sizeof(kb));
		memset(&kb.kb_freelist, 0, sizeof(kb.kb_freelist));
		return (sysctl_rdstruct(oldp, oldlenp, newp, &kb, sizeof(kb)));
	case KERN_MALLOC_KMEMSTATS:
#ifdef KMEMSTATS
		if ((name[1] < 0) || (name[1] >= M_LAST))
			return (EINVAL);
		return (sysctl_rdstruct(oldp, oldlenp, newp,
		    &kmemstats[name[1]], sizeof(struct kmemstats)));
#else
		return (EOPNOTSUPP);
#endif
	case KERN_MALLOC_KMEMNAMES:
#if defined(KMEMSTATS) || defined(DIAGNOSTIC) || defined(FFS_SOFTUPDATES)
		error = rw_enter(&sysctl_kmemlock, RW_WRITE|RW_INTR);
		if (error)
			return (error);
		if (memall == NULL) {
			int totlen;

			/* Figure out how large a buffer we need */
			for (totlen = 0, i = 0; i < M_LAST; i++) {
				if (memname[i])
					totlen += strlen(memname[i]);
				totlen++;
			}
			memall = malloc(totlen + M_LAST, M_SYSCTL,
			    M_WAITOK|M_ZERO);
			for (siz = 0, i = 0; i < M_LAST; i++) {
				snprintf(memall + siz, 
				    totlen + M_LAST - siz,
				    "%s,", memname[i] ? memname[i] : "");
				siz += strlen(memall + siz);
			}
			/* Remove trailing comma */
			if (siz)
				memall[siz - 1] = '\0';

			/* Now, convert all spaces to underscores */
			for (i = 0; i < totlen; i++)
				if (memall[i] == ' ')
					memall[i] = '_';
		}
		rw_exit_write(&sysctl_kmemlock);
		return (sysctl_rdstring(oldp, oldlenp, newp, memall));
#else
		return (EOPNOTSUPP);
#endif
	default:
		return (EOPNOTSUPP);
	}
	/* NOTREACHED */
}

/*
 * Round up a size to how much malloc would actually allocate.
 */
size_t
malloc_roundup(size_t sz)
{
	if (sz > MAXALLOCSAVE)
		return round_page(sz);

	return (1 << BUCKETINDX(sz));
}

#if defined(DDB)
#include <machine/db_machdep.h>
#include <ddb/db_output.h>

void
malloc_printit(
    int (*pr)(const char *, ...) __attribute__((__format__(__kprintf__,1,2))))
{
#ifdef KMEMSTATS
	struct kmemstats *km;
	int i;

	(*pr)("%15s %5s  %6s  %7s  %6s %9s %8s %8s\n",
	    "Type", "InUse", "MemUse", "HighUse", "Limit", "Requests",
	    "Type Lim", "Kern Lim");
	for (i = 0, km = kmemstats; i < M_LAST; i++, km++) {
		if (!km->ks_calls || !memname[i])
			continue;

		(*pr)("%15s %5ld %6ldK %7ldK %6ldK %9ld %8d %8d\n",
		    memname[i], km->ks_inuse, km->ks_memuse / 1024,
		    km->ks_maxused / 1024, km->ks_limit / 1024, 
		    km->ks_calls, km->ks_limblocks, km->ks_mapblocks);
	}
#else
	(*pr)("No KMEMSTATS compiled in\n");
#endif
}
#endif /* DDB */

/*
 * Copyright (c) 2008 Otto Moerbeek <otto@drijf.net>
 *
 * Permission to use, copy, modify, and distribute this software for any
 * purpose with or without fee is hereby granted, provided that the above
 * copyright notice and this permission notice appear in all copies.
 *
 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 */

/*
 * This is sqrt(SIZE_MAX+1), as s1*s2 <= SIZE_MAX
 * if both s1 < MUL_NO_OVERFLOW and s2 < MUL_NO_OVERFLOW
 */
#define MUL_NO_OVERFLOW	(1UL << (sizeof(size_t) * 4))

void *
mallocarray(size_t nmemb, size_t size, int type, int flags)
{
	if ((nmemb >= MUL_NO_OVERFLOW || size >= MUL_NO_OVERFLOW) &&
	    nmemb > 0 && SIZE_MAX / nmemb < size) {
		if (flags & M_CANFAIL)
			return (NULL);
		panic("mallocarray: overflow %zu * %zu", nmemb, size);
	}
	return (malloc(size * nmemb, type, flags));
}