summaryrefslogtreecommitdiff
path: root/lib/librte_eal/common/eal_common_memory.c
blob: 3e30c58445908570ab8686773ebbb98ea2d500ee (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
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
/* SPDX-License-Identifier: BSD-3-Clause
 * Copyright(c) 2010-2014 Intel Corporation
 */

#include <errno.h>
#include <stdio.h>
#include <stdint.h>
#include <stdlib.h>
#include <stdarg.h>
#include <string.h>
#include <unistd.h>
#include <inttypes.h>
#include <sys/mman.h>
#include <sys/queue.h>

#include <rte_fbarray.h>
#include <rte_memory.h>
#include <rte_eal.h>
#include <rte_eal_memconfig.h>
#include <rte_errno.h>
#include <rte_log.h>

#include "eal_memalloc.h"
#include "eal_private.h"
#include "eal_internal_cfg.h"

/*
 * Try to mmap *size bytes in /dev/zero. If it is successful, return the
 * pointer to the mmap'd area and keep *size unmodified. Else, retry
 * with a smaller zone: decrease *size by hugepage_sz until it reaches
 * 0. In this case, return NULL. Note: this function returns an address
 * which is a multiple of hugepage size.
 */

#define MEMSEG_LIST_FMT "memseg-%" PRIu64 "k-%i-%i"

static uint64_t baseaddr_offset;
static uint64_t system_page_sz;

void *
eal_get_virtual_area(void *requested_addr, size_t *size,
		size_t page_sz, int flags, int mmap_flags)
{
	bool addr_is_hint, allow_shrink, unmap, no_align;
	uint64_t map_sz;
	void *mapped_addr, *aligned_addr;

	if (system_page_sz == 0)
		system_page_sz = sysconf(_SC_PAGESIZE);

	mmap_flags |= MAP_PRIVATE | MAP_ANONYMOUS;

	RTE_LOG(DEBUG, EAL, "Ask a virtual area of 0x%zx bytes\n", *size);

	addr_is_hint = (flags & EAL_VIRTUAL_AREA_ADDR_IS_HINT) > 0;
	allow_shrink = (flags & EAL_VIRTUAL_AREA_ALLOW_SHRINK) > 0;
	unmap = (flags & EAL_VIRTUAL_AREA_UNMAP) > 0;

	if (requested_addr == NULL && internal_config.base_virtaddr != 0) {
		requested_addr = (void *) (internal_config.base_virtaddr +
				(size_t)baseaddr_offset);
		requested_addr = RTE_PTR_ALIGN(requested_addr, page_sz);
		addr_is_hint = true;
	}

	/* if requested address is not aligned by page size, or if requested
	 * address is NULL, add page size to requested length as we may get an
	 * address that's aligned by system page size, which can be smaller than
	 * our requested page size. additionally, we shouldn't try to align if
	 * system page size is the same as requested page size.
	 */
	no_align = (requested_addr != NULL &&
		((uintptr_t)requested_addr & (page_sz - 1)) == 0) ||
		page_sz == system_page_sz;

	do {
		map_sz = no_align ? *size : *size + page_sz;

		mapped_addr = mmap(requested_addr, map_sz, PROT_READ,
				mmap_flags, -1, 0);
		if (mapped_addr == MAP_FAILED && allow_shrink)
			*size -= page_sz;
	} while (allow_shrink && mapped_addr == MAP_FAILED && *size > 0);

	/* align resulting address - if map failed, we will ignore the value
	 * anyway, so no need to add additional checks.
	 */
	aligned_addr = no_align ? mapped_addr :
			RTE_PTR_ALIGN(mapped_addr, page_sz);

	if (*size == 0) {
		RTE_LOG(ERR, EAL, "Cannot get a virtual area of any size: %s\n",
			strerror(errno));
		rte_errno = errno;
		return NULL;
	} else if (mapped_addr == MAP_FAILED) {
		RTE_LOG(ERR, EAL, "Cannot get a virtual area: %s\n",
			strerror(errno));
		/* pass errno up the call chain */
		rte_errno = errno;
		return NULL;
	} else if (requested_addr != NULL && !addr_is_hint &&
			aligned_addr != requested_addr) {
		RTE_LOG(ERR, EAL, "Cannot get a virtual area at requested address: %p (got %p)\n",
			requested_addr, aligned_addr);
		munmap(mapped_addr, map_sz);
		rte_errno = EADDRNOTAVAIL;
		return NULL;
	} else if (requested_addr != NULL && addr_is_hint &&
			aligned_addr != requested_addr) {
		RTE_LOG(WARNING, EAL, "WARNING! Base virtual address hint (%p != %p) not respected!\n",
			requested_addr, aligned_addr);
		RTE_LOG(WARNING, EAL, "   This may cause issues with mapping memory into secondary processes\n");
	}

	RTE_LOG(DEBUG, EAL, "Virtual area found at %p (size = 0x%zx)\n",
		aligned_addr, *size);

	if (unmap)
		munmap(mapped_addr, map_sz);

	baseaddr_offset += *size;

	return aligned_addr;
}

static uint64_t
get_mem_amount(uint64_t page_sz, uint64_t max_mem)
{
	uint64_t area_sz, max_pages;

	/* limit to RTE_MAX_MEMSEG_PER_LIST pages or RTE_MAX_MEM_MB_PER_LIST */
	max_pages = RTE_MAX_MEMSEG_PER_LIST;
	max_mem = RTE_MIN((uint64_t)RTE_MAX_MEM_MB_PER_LIST << 20, max_mem);

	area_sz = RTE_MIN(page_sz * max_pages, max_mem);

	/* make sure the list isn't smaller than the page size */
	area_sz = RTE_MAX(area_sz, page_sz);

	return RTE_ALIGN(area_sz, page_sz);
}

static int
alloc_memseg_list(struct rte_memseg_list *msl, uint64_t page_sz,
		uint64_t max_mem, int socket_id, int type_msl_idx)
{
	char name[RTE_FBARRAY_NAME_LEN];
	uint64_t mem_amount;
	int max_segs;

	mem_amount = get_mem_amount(page_sz, max_mem);
	max_segs = mem_amount / page_sz;

	snprintf(name, sizeof(name), MEMSEG_LIST_FMT, page_sz >> 10, socket_id,
		 type_msl_idx);
	if (rte_fbarray_init(&msl->memseg_arr, name, max_segs,
			sizeof(struct rte_memseg))) {
		RTE_LOG(ERR, EAL, "Cannot allocate memseg list: %s\n",
			rte_strerror(rte_errno));
		return -1;
	}

	msl->page_sz = page_sz;
	msl->socket_id = socket_id;
	msl->base_va = NULL;

	RTE_LOG(DEBUG, EAL, "Memseg list allocated: 0x%zxkB at socket %i\n",
			(size_t)page_sz >> 10, socket_id);

	return 0;
}

static int
alloc_va_space(struct rte_memseg_list *msl)
{
	uint64_t page_sz;
	size_t mem_sz;
	void *addr;
	int flags = 0;

#ifdef RTE_ARCH_PPC_64
	flags |= MAP_HUGETLB;
#endif

	page_sz = msl->page_sz;
	mem_sz = page_sz * msl->memseg_arr.len;

	addr = eal_get_virtual_area(msl->base_va, &mem_sz, page_sz, 0, flags);
	if (addr == NULL) {
		if (rte_errno == EADDRNOTAVAIL)
			RTE_LOG(ERR, EAL, "Could not mmap %llu bytes at [%p] - please use '--base-virtaddr' option\n",
				(unsigned long long)mem_sz, msl->base_va);
		else
			RTE_LOG(ERR, EAL, "Cannot reserve memory\n");
		return -1;
	}
	msl->base_va = addr;

	return 0;
}

static int __rte_unused
memseg_primary_init_32(void)
{
	struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
	int active_sockets, hpi_idx, msl_idx = 0;
	unsigned int socket_id, i;
	struct rte_memseg_list *msl;
	uint64_t extra_mem_per_socket, total_extra_mem, total_requested_mem;
	uint64_t max_mem;

	/* no-huge does not need this at all */
	if (internal_config.no_hugetlbfs)
		return 0;

	/* this is a giant hack, but desperate times call for desperate
	 * measures. in legacy 32-bit mode, we cannot preallocate VA space,
	 * because having upwards of 2 gigabytes of VA space already mapped will
	 * interfere with our ability to map and sort hugepages.
	 *
	 * therefore, in legacy 32-bit mode, we will be initializing memseg
	 * lists much later - in eal_memory.c, right after we unmap all the
	 * unneeded pages. this will not affect secondary processes, as those
	 * should be able to mmap the space without (too many) problems.
	 */
	if (internal_config.legacy_mem)
		return 0;

	/* 32-bit mode is a very special case. we cannot know in advance where
	 * the user will want to allocate their memory, so we have to do some
	 * heuristics.
	 */
	active_sockets = 0;
	total_requested_mem = 0;
	if (internal_config.force_sockets)
		for (i = 0; i < rte_socket_count(); i++) {
			uint64_t mem;

			socket_id = rte_socket_id_by_idx(i);
			mem = internal_config.socket_mem[socket_id];

			if (mem == 0)
				continue;

			active_sockets++;
			total_requested_mem += mem;
		}
	else
		total_requested_mem = internal_config.memory;

	max_mem = (uint64_t) RTE_MAX_MEM_MB_PER_TYPE << 20;
	if (total_requested_mem > max_mem) {
		RTE_LOG(ERR, EAL, "Invalid parameters: 32-bit process can at most use %uM of memory\n",
				(unsigned int)(max_mem >> 20));
		return -1;
	}
	total_extra_mem = max_mem - total_requested_mem;
	extra_mem_per_socket = active_sockets == 0 ? total_extra_mem :
			total_extra_mem / active_sockets;

	/* the allocation logic is a little bit convoluted, but here's how it
	 * works, in a nutshell:
	 *  - if user hasn't specified on which sockets to allocate memory via
	 *    --socket-mem, we allocate all of our memory on master core socket.
	 *  - if user has specified sockets to allocate memory on, there may be
	 *    some "unused" memory left (e.g. if user has specified --socket-mem
	 *    such that not all memory adds up to 2 gigabytes), so add it to all
	 *    sockets that are in use equally.
	 *
	 * page sizes are sorted by size in descending order, so we can safely
	 * assume that we dispense with bigger page sizes first.
	 */

	/* create memseg lists */
	for (i = 0; i < rte_socket_count(); i++) {
		int hp_sizes = (int) internal_config.num_hugepage_sizes;
		uint64_t max_socket_mem, cur_socket_mem;
		unsigned int master_lcore_socket;
		struct rte_config *cfg = rte_eal_get_configuration();
		bool skip;

		socket_id = rte_socket_id_by_idx(i);

#ifndef RTE_EAL_NUMA_AWARE_HUGEPAGES
		if (socket_id > 0)
			break;
#endif

		/* if we didn't specifically request memory on this socket */
		skip = active_sockets != 0 &&
				internal_config.socket_mem[socket_id] == 0;
		/* ...or if we didn't specifically request memory on *any*
		 * socket, and this is not master lcore
		 */
		master_lcore_socket = rte_lcore_to_socket_id(cfg->master_lcore);
		skip |= active_sockets == 0 && socket_id != master_lcore_socket;

		if (skip) {
			RTE_LOG(DEBUG, EAL, "Will not preallocate memory on socket %u\n",
					socket_id);
			continue;
		}

		/* max amount of memory on this socket */
		max_socket_mem = (active_sockets != 0 ?
					internal_config.socket_mem[socket_id] :
					internal_config.memory) +
					extra_mem_per_socket;
		cur_socket_mem = 0;

		for (hpi_idx = 0; hpi_idx < hp_sizes; hpi_idx++) {
			uint64_t max_pagesz_mem, cur_pagesz_mem = 0;
			uint64_t hugepage_sz;
			struct hugepage_info *hpi;
			int type_msl_idx, max_segs, total_segs = 0;

			hpi = &internal_config.hugepage_info[hpi_idx];
			hugepage_sz = hpi->hugepage_sz;

			max_segs = RTE_MAX_MEMSEG_PER_TYPE;
			max_pagesz_mem = max_socket_mem - cur_socket_mem;

			/* make it multiple of page size */
			max_pagesz_mem = RTE_ALIGN_FLOOR(max_pagesz_mem,
					hugepage_sz);

			RTE_LOG(DEBUG, EAL, "Attempting to preallocate "
					"%" PRIu64 "M on socket %i\n",
					max_pagesz_mem >> 20, socket_id);

			type_msl_idx = 0;
			while (cur_pagesz_mem < max_pagesz_mem &&
					total_segs < max_segs) {
				if (msl_idx >= RTE_MAX_MEMSEG_LISTS) {
					RTE_LOG(ERR, EAL,
						"No more space in memseg lists, please increase %s\n",
						RTE_STR(CONFIG_RTE_MAX_MEMSEG_LISTS));
					return -1;
				}

				msl = &mcfg->memsegs[msl_idx++];

				if (alloc_memseg_list(msl, hugepage_sz,
						max_pagesz_mem, socket_id,
						type_msl_idx))
					return -1;

				total_segs += msl->memseg_arr.len;
				cur_pagesz_mem = total_segs * hugepage_sz;
				type_msl_idx++;

				if (alloc_va_space(msl)) {
					RTE_LOG(ERR, EAL, "Cannot allocate VA space for memseg list\n");
					return -1;
				}
			}
			cur_socket_mem += cur_pagesz_mem;
		}
	}

	return 0;
}

static int __rte_unused
memseg_primary_init(void)
{
	struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
	int i, socket_id, hpi_idx, msl_idx = 0;
	struct rte_memseg_list *msl;
	uint64_t max_mem, total_mem;

	/* no-huge does not need this at all */
	if (internal_config.no_hugetlbfs)
		return 0;

	max_mem = (uint64_t)RTE_MAX_MEM_MB << 20;
	total_mem = 0;

	/* create memseg lists */
	for (hpi_idx = 0; hpi_idx < (int) internal_config.num_hugepage_sizes;
			hpi_idx++) {
		struct hugepage_info *hpi;
		uint64_t hugepage_sz;

		hpi = &internal_config.hugepage_info[hpi_idx];
		hugepage_sz = hpi->hugepage_sz;

		for (i = 0; i < (int) rte_socket_count(); i++) {
			uint64_t max_type_mem, total_type_mem = 0;
			int type_msl_idx, max_segs, total_segs = 0;

			socket_id = rte_socket_id_by_idx(i);

#ifndef RTE_EAL_NUMA_AWARE_HUGEPAGES
			if (socket_id > 0)
				break;
#endif

			max_type_mem = RTE_MIN(max_mem - total_mem,
				(uint64_t)RTE_MAX_MEM_MB_PER_TYPE << 20);
			max_segs = RTE_MAX_MEMSEG_PER_TYPE;

			type_msl_idx = 0;
			while (total_type_mem < max_type_mem &&
					total_segs < max_segs) {
				uint64_t cur_max_mem;
				if (msl_idx >= RTE_MAX_MEMSEG_LISTS) {
					RTE_LOG(ERR, EAL,
						"No more space in memseg lists, please increase %s\n",
						RTE_STR(CONFIG_RTE_MAX_MEMSEG_LISTS));
					return -1;
				}

				msl = &mcfg->memsegs[msl_idx++];

				cur_max_mem = max_type_mem - total_type_mem;
				if (alloc_memseg_list(msl, hugepage_sz,
						cur_max_mem, socket_id,
						type_msl_idx))
					return -1;

				total_segs += msl->memseg_arr.len;
				total_type_mem = total_segs * hugepage_sz;
				type_msl_idx++;

				if (alloc_va_space(msl)) {
					RTE_LOG(ERR, EAL, "Cannot allocate VA space for memseg list\n");
					return -1;
				}
			}
			total_mem += total_type_mem;
		}
	}
	return 0;
}

static int
memseg_secondary_init(void)
{
	struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
	int msl_idx = 0;
	struct rte_memseg_list *msl;

	for (msl_idx = 0; msl_idx < RTE_MAX_MEMSEG_LISTS; msl_idx++) {

		msl = &mcfg->memsegs[msl_idx];

		/* skip empty memseg lists */
		if (msl->memseg_arr.len == 0)
			continue;

		if (rte_fbarray_attach(&msl->memseg_arr)) {
			RTE_LOG(ERR, EAL, "Cannot attach to primary process memseg lists\n");
			return -1;
		}

		/* preallocate VA space */
		if (alloc_va_space(msl)) {
			RTE_LOG(ERR, EAL, "Cannot preallocate VA space for hugepage memory\n");
			return -1;
		}
	}

	return 0;
}

static struct rte_memseg *
virt2memseg(const void *addr, const struct rte_memseg_list *msl)
{
	const struct rte_fbarray *arr;
	void *start, *end;
	int ms_idx;

	/* a memseg list was specified, check if it's the right one */
	start = msl->base_va;
	end = RTE_PTR_ADD(start, (size_t)msl->page_sz * msl->memseg_arr.len);

	if (addr < start || addr >= end)
		return NULL;

	/* now, calculate index */
	arr = &msl->memseg_arr;
	ms_idx = RTE_PTR_DIFF(addr, msl->base_va) / msl->page_sz;
	return rte_fbarray_get(arr, ms_idx);
}

static struct rte_memseg_list *
virt2memseg_list(const void *addr)
{
	struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
	struct rte_memseg_list *msl;
	int msl_idx;

	for (msl_idx = 0; msl_idx < RTE_MAX_MEMSEG_LISTS; msl_idx++) {
		void *start, *end;
		msl = &mcfg->memsegs[msl_idx];

		start = msl->base_va;
		end = RTE_PTR_ADD(start,
				(size_t)msl->page_sz * msl->memseg_arr.len);
		if (addr >= start && addr < end)
			break;
	}
	/* if we didn't find our memseg list */
	if (msl_idx == RTE_MAX_MEMSEG_LISTS)
		return NULL;
	return msl;
}

__rte_experimental struct rte_memseg_list *
rte_mem_virt2memseg_list(const void *addr)
{
	return virt2memseg_list(addr);
}

struct virtiova {
	rte_iova_t iova;
	void *virt;
};
static int
find_virt(const struct rte_memseg_list *msl __rte_unused,
		const struct rte_memseg *ms, void *arg)
{
	struct virtiova *vi = arg;
	if (vi->iova >= ms->iova && vi->iova < (ms->iova + ms->len)) {
		size_t offset = vi->iova - ms->iova;
		vi->virt = RTE_PTR_ADD(ms->addr, offset);
		/* stop the walk */
		return 1;
	}
	return 0;
}
static int
find_virt_legacy(const struct rte_memseg_list *msl __rte_unused,
		const struct rte_memseg *ms, size_t len, void *arg)
{
	struct virtiova *vi = arg;
	if (vi->iova >= ms->iova && vi->iova < (ms->iova + len)) {
		size_t offset = vi->iova - ms->iova;
		vi->virt = RTE_PTR_ADD(ms->addr, offset);
		/* stop the walk */
		return 1;
	}
	return 0;
}

__rte_experimental void *
rte_mem_iova2virt(rte_iova_t iova)
{
	struct virtiova vi;

	memset(&vi, 0, sizeof(vi));

	vi.iova = iova;
	/* for legacy mem, we can get away with scanning VA-contiguous segments,
	 * as we know they are PA-contiguous as well
	 */
	if (internal_config.legacy_mem)
		rte_memseg_contig_walk(find_virt_legacy, &vi);
	else
		rte_memseg_walk(find_virt, &vi);

	return vi.virt;
}

__rte_experimental struct rte_memseg *
rte_mem_virt2memseg(const void *addr, const struct rte_memseg_list *msl)
{
	return virt2memseg(addr, msl != NULL ? msl :
			rte_mem_virt2memseg_list(addr));
}

static int
physmem_size(const struct rte_memseg_list *msl, void *arg)
{
	uint64_t *total_len = arg;

	*total_len += msl->memseg_arr.count * msl->page_sz;

	return 0;
}

/* get the total size of memory */
uint64_t
rte_eal_get_physmem_size(void)
{
	uint64_t total_len = 0;

	rte_memseg_list_walk(physmem_size, &total_len);

	return total_len;
}

static int
dump_memseg(const struct rte_memseg_list *msl, const struct rte_memseg *ms,
		void *arg)
{
	struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
	int msl_idx, ms_idx;
	FILE *f = arg;

	msl_idx = msl - mcfg->memsegs;
	if (msl_idx < 0 || msl_idx >= RTE_MAX_MEMSEG_LISTS)
		return -1;

	ms_idx = rte_fbarray_find_idx(&msl->memseg_arr, ms);
	if (ms_idx < 0)
		return -1;

	fprintf(f, "Segment %i-%i: IOVA:0x%"PRIx64", len:%zu, "
			"virt:%p, socket_id:%"PRId32", "
			"hugepage_sz:%"PRIu64", nchannel:%"PRIx32", "
			"nrank:%"PRIx32"\n",
			msl_idx, ms_idx,
			ms->iova,
			ms->len,
			ms->addr,
			ms->socket_id,
			ms->hugepage_sz,
			ms->nchannel,
			ms->nrank);

	return 0;
}

/*
 * Defining here because declared in rte_memory.h, but the actual implementation
 * is in eal_common_memalloc.c, like all other memalloc internals.
 */
int __rte_experimental
rte_mem_event_callback_register(const char *name, rte_mem_event_callback_t clb)
{
	/* FreeBSD boots with legacy mem enabled by default */
	if (internal_config.legacy_mem) {
		RTE_LOG(DEBUG, EAL, "Registering mem event callbacks not supported\n");
		rte_errno = ENOTSUP;
		return -1;
	}
	return eal_memalloc_mem_event_callback_register(name, clb);
}

int __rte_experimental
rte_mem_event_callback_unregister(const char *name)
{
	/* FreeBSD boots with legacy mem enabled by default */
	if (internal_config.legacy_mem) {
		RTE_LOG(DEBUG, EAL, "Registering mem event callbacks not supported\n");
		rte_errno = ENOTSUP;
		return -1;
	}
	return eal_memalloc_mem_event_callback_unregister(name);
}

int __rte_experimental
rte_mem_alloc_validator_register(const char *name,
		rte_mem_alloc_validator_t clb, int socket_id, size_t limit)
{
	/* FreeBSD boots with legacy mem enabled by default */
	if (internal_config.legacy_mem) {
		RTE_LOG(DEBUG, EAL, "Registering mem alloc validators not supported\n");
		rte_errno = ENOTSUP;
		return -1;
	}
	return eal_memalloc_mem_alloc_validator_register(name, clb, socket_id,
			limit);
}

int __rte_experimental
rte_mem_alloc_validator_unregister(const char *name, int socket_id)
{
	/* FreeBSD boots with legacy mem enabled by default */
	if (internal_config.legacy_mem) {
		RTE_LOG(DEBUG, EAL, "Registering mem alloc validators not supported\n");
		rte_errno = ENOTSUP;
		return -1;
	}
	return eal_memalloc_mem_alloc_validator_unregister(name, socket_id);
}

/* Dump the physical memory layout on console */
void
rte_dump_physmem_layout(FILE *f)
{
	rte_memseg_walk(dump_memseg, f);
}

/* return the number of memory channels */
unsigned rte_memory_get_nchannel(void)
{
	return rte_eal_get_configuration()->mem_config->nchannel;
}

/* return the number of memory rank */
unsigned rte_memory_get_nrank(void)
{
	return rte_eal_get_configuration()->mem_config->nrank;
}

static int
rte_eal_memdevice_init(void)
{
	struct rte_config *config;

	if (rte_eal_process_type() == RTE_PROC_SECONDARY)
		return 0;

	config = rte_eal_get_configuration();
	config->mem_config->nchannel = internal_config.force_nchannel;
	config->mem_config->nrank = internal_config.force_nrank;

	return 0;
}

/* Lock page in physical memory and prevent from swapping. */
int
rte_mem_lock_page(const void *virt)
{
	unsigned long virtual = (unsigned long)virt;
	int page_size = getpagesize();
	unsigned long aligned = (virtual & ~(page_size - 1));
	return mlock((void *)aligned, page_size);
}

int __rte_experimental
rte_memseg_contig_walk(rte_memseg_contig_walk_t func, void *arg)
{
	struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
	int i, ms_idx, ret = 0;

	/* do not allow allocations/frees/init while we iterate */
	rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);

	for (i = 0; i < RTE_MAX_MEMSEG_LISTS; i++) {
		struct rte_memseg_list *msl = &mcfg->memsegs[i];
		const struct rte_memseg *ms;
		struct rte_fbarray *arr;

		if (msl->memseg_arr.count == 0)
			continue;

		arr = &msl->memseg_arr;

		ms_idx = rte_fbarray_find_next_used(arr, 0);
		while (ms_idx >= 0) {
			int n_segs;
			size_t len;

			ms = rte_fbarray_get(arr, ms_idx);

			/* find how many more segments there are, starting with
			 * this one.
			 */
			n_segs = rte_fbarray_find_contig_used(arr, ms_idx);
			len = n_segs * msl->page_sz;

			ret = func(msl, ms, len, arg);
			if (ret < 0) {
				ret = -1;
				goto out;
			} else if (ret > 0) {
				ret = 1;
				goto out;
			}
			ms_idx = rte_fbarray_find_next_used(arr,
					ms_idx + n_segs);
		}
	}
out:
	rte_rwlock_read_unlock(&mcfg->memory_hotplug_lock);
	return ret;
}

int __rte_experimental
rte_memseg_walk(rte_memseg_walk_t func, void *arg)
{
	struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
	int i, ms_idx, ret = 0;

	/* do not allow allocations/frees/init while we iterate */
	rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);

	for (i = 0; i < RTE_MAX_MEMSEG_LISTS; i++) {
		struct rte_memseg_list *msl = &mcfg->memsegs[i];
		const struct rte_memseg *ms;
		struct rte_fbarray *arr;

		if (msl->memseg_arr.count == 0)
			continue;

		arr = &msl->memseg_arr;

		ms_idx = rte_fbarray_find_next_used(arr, 0);
		while (ms_idx >= 0) {
			ms = rte_fbarray_get(arr, ms_idx);
			ret = func(msl, ms, arg);
			if (ret < 0) {
				ret = -1;
				goto out;
			} else if (ret > 0) {
				ret = 1;
				goto out;
			}
			ms_idx = rte_fbarray_find_next_used(arr, ms_idx + 1);
		}
	}
out:
	rte_rwlock_read_unlock(&mcfg->memory_hotplug_lock);
	return ret;
}

int __rte_experimental
rte_memseg_list_walk(rte_memseg_list_walk_t func, void *arg)
{
	struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
	int i, ret = 0;

	/* do not allow allocations/frees/init while we iterate */
	rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);

	for (i = 0; i < RTE_MAX_MEMSEG_LISTS; i++) {
		struct rte_memseg_list *msl = &mcfg->memsegs[i];

		if (msl->base_va == NULL)
			continue;

		ret = func(msl, arg);
		if (ret < 0) {
			ret = -1;
			goto out;
		}
		if (ret > 0) {
			ret = 1;
			goto out;
		}
	}
out:
	rte_rwlock_read_unlock(&mcfg->memory_hotplug_lock);
	return ret;
}

/* init memory subsystem */
int
rte_eal_memory_init(void)
{
	struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
	int retval;
	RTE_LOG(DEBUG, EAL, "Setting up physically contiguous memory...\n");

	if (!mcfg)
		return -1;

	/* lock mem hotplug here, to prevent races while we init */
	rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);

	retval = rte_eal_process_type() == RTE_PROC_PRIMARY ?
#ifndef RTE_ARCH_64
			memseg_primary_init_32() :
#else
			memseg_primary_init() :
#endif
			memseg_secondary_init();

	if (retval < 0)
		goto fail;

	retval = rte_eal_process_type() == RTE_PROC_PRIMARY ?
			rte_eal_hugepage_init() :
			rte_eal_hugepage_attach();
	if (retval < 0)
		goto fail;

	if (internal_config.no_shconf == 0 && rte_eal_memdevice_init() < 0)
		goto fail;

	return 0;
fail:
	rte_rwlock_read_unlock(&mcfg->memory_hotplug_lock);
	return -1;
}