DPDK logo

Elixir Cross Referencer

  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
/* 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 void *next_baseaddr;
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 (next_baseaddr == NULL && internal_config.base_virtaddr != 0 &&
			rte_eal_process_type() == RTE_PROC_PRIMARY)
		next_baseaddr = (void *) internal_config.base_virtaddr;

	if (requested_addr == NULL && next_baseaddr != NULL) {
		requested_addr = next_baseaddr;
		requested_addr = RTE_PTR_ALIGN(requested_addr, page_sz);
		addr_is_hint = true;
	}

	/* we don't need alignment of resulting pointer in the following cases:
	 *
	 * 1. page size is equal to system size
	 * 2. we have a requested address, and it is page-aligned, and we will
	 *    be discarding the address if we get a different one.
	 *
	 * for all other cases, alignment is potentially necessary.
	 */
	no_align = (requested_addr != NULL &&
		requested_addr == RTE_PTR_ALIGN(requested_addr, page_sz) &&
		!addr_is_hint) ||
		page_sz == system_page_sz;

	do {
		map_sz = no_align ? *size : *size + page_sz;
		if (map_sz > SIZE_MAX) {
			RTE_LOG(ERR, EAL, "Map size too big\n");
			rte_errno = E2BIG;
			return NULL;
		}

		mapped_addr = mmap(requested_addr, (size_t)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");
	} else if (next_baseaddr != NULL) {
		next_baseaddr = RTE_PTR_ADD(aligned_addr, *size);
	}

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

	if (unmap) {
		munmap(mapped_addr, map_sz);
	} else if (!no_align) {
		void *map_end, *aligned_end;
		size_t before_len, after_len;

		/* when we reserve space with alignment, we add alignment to
		 * mapping size. On 32-bit, if 1GB alignment was requested, this
		 * would waste 1GB of address space, which is a luxury we cannot
		 * afford. so, if alignment was performed, check if any unneeded
		 * address space can be unmapped back.
		 */

		map_end = RTE_PTR_ADD(mapped_addr, (size_t)map_sz);
		aligned_end = RTE_PTR_ADD(aligned_addr, *size);

		/* unmap space before aligned mmap address */
		before_len = RTE_PTR_DIFF(aligned_addr, mapped_addr);
		if (before_len > 0)
			munmap(mapped_addr, before_len);

		/* unmap space after aligned end mmap address */
		after_len = RTE_PTR_DIFF(map_end, aligned_end);
		if (after_len > 0)
			munmap(aligned_end, after_len);
	}

	return aligned_addr;
}

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;

	if (msl == NULL)
		return NULL;

	/* 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,
		void *arg)
{
	/* 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, arg);
}

int __rte_experimental
rte_mem_event_callback_unregister(const char *name, void *arg)
{
	/* 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, arg);
}

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_thread_unsafe(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;

	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)
				return ret;
			ms_idx = rte_fbarray_find_next_used(arr,
					ms_idx + n_segs);
		}
	}
	return 0;
}

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 ret = 0;

	/* do not allow allocations/frees/init while we iterate */
	rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);
	ret = rte_memseg_contig_walk_thread_unsafe(func, arg);
	rte_rwlock_read_unlock(&mcfg->memory_hotplug_lock);

	return ret;
}

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

	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)
				return ret;
			ms_idx = rte_fbarray_find_next_used(arr, ms_idx + 1);
		}
	}
	return 0;
}

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 ret = 0;

	/* do not allow allocations/frees/init while we iterate */
	rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);
	ret = rte_memseg_walk_thread_unsafe(func, arg);
	rte_rwlock_read_unlock(&mcfg->memory_hotplug_lock);

	return ret;
}

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

	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)
			return ret;
	}
	return 0;
}

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 ret = 0;

	/* do not allow allocations/frees/init while we iterate */
	rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);
	ret = rte_memseg_list_walk_thread_unsafe(func, arg);
	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);

	if (rte_eal_memseg_init() < 0)
		goto fail;

	if (eal_memalloc_init() < 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;
}