aboutsummaryrefslogtreecommitdiffstats
path: root/arch/x86/mm/numa_64.c
blob: 71a14f89f89e7367c72f2de85da6708acc83c069 (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
/*
 * Generic VM initialization for x86-64 NUMA setups.
 * Copyright 2002,2003 Andi Kleen, SuSE Labs.
 */
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/string.h>
#include <linux/init.h>
#include <linux/bootmem.h>
#include <linux/mmzone.h>
#include <linux/ctype.h>
#include <linux/module.h>
#include <linux/nodemask.h>
#include <linux/sched.h>

#include <asm/e820.h>
#include <asm/proto.h>
#include <asm/dma.h>
#include <asm/numa.h>
#include <asm/acpi.h>
#include <asm/k8.h>

struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
EXPORT_SYMBOL(node_data);

struct memnode memnode;

s16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
};

int numa_off __initdata;
static unsigned long __initdata nodemap_addr;
static unsigned long __initdata nodemap_size;

/*
 * Given a shift value, try to populate memnodemap[]
 * Returns :
 * 1 if OK
 * 0 if memnodmap[] too small (of shift too small)
 * -1 if node overlap or lost ram (shift too big)
 */
static int __init populate_memnodemap(const struct bootnode *nodes,
				      int numnodes, int shift, int *nodeids)
{
	unsigned long addr, end;
	int i, res = -1;

	memset(memnodemap, 0xff, sizeof(s16)*memnodemapsize);
	for (i = 0; i < numnodes; i++) {
		addr = nodes[i].start;
		end = nodes[i].end;
		if (addr >= end)
			continue;
		if ((end >> shift) >= memnodemapsize)
			return 0;
		do {
			if (memnodemap[addr >> shift] != NUMA_NO_NODE)
				return -1;

			if (!nodeids)
				memnodemap[addr >> shift] = i;
			else
				memnodemap[addr >> shift] = nodeids[i];

			addr += (1UL << shift);
		} while (addr < end);
		res = 1;
	}
	return res;
}

static int __init allocate_cachealigned_memnodemap(void)
{
	unsigned long addr;

	memnodemap = memnode.embedded_map;
	if (memnodemapsize <= ARRAY_SIZE(memnode.embedded_map))
		return 0;

	addr = 0x8000;
	nodemap_size = roundup(sizeof(s16) * memnodemapsize, L1_CACHE_BYTES);
	nodemap_addr = find_e820_area(addr, max_pfn<<PAGE_SHIFT,
				      nodemap_size, L1_CACHE_BYTES);
	if (nodemap_addr == -1UL) {
		printk(KERN_ERR
		       "NUMA: Unable to allocate Memory to Node hash map\n");
		nodemap_addr = nodemap_size = 0;
		return -1;
	}
	memnodemap = phys_to_virt(nodemap_addr);
	reserve_early(nodemap_addr, nodemap_addr + nodemap_size, "MEMNODEMAP");

	printk(KERN_DEBUG "NUMA: Allocated memnodemap from %lx - %lx\n",
	       nodemap_addr, nodemap_addr + nodemap_size);
	return 0;
}

/*
 * The LSB of all start and end addresses in the node map is the value of the
 * maximum possible shift.
 */
static int __init extract_lsb_from_nodes(const struct bootnode *nodes,
					 int numnodes)
{
	int i, nodes_used = 0;
	unsigned long start, end;
	unsigned long bitfield = 0, memtop = 0;

	for (i = 0; i < numnodes; i++) {
		start = nodes[i].start;
		end = nodes[i].end;
		if (start >= end)
			continue;
		bitfield |= start;
		nodes_used++;
		if (end > memtop)
			memtop = end;
	}
	if (nodes_used <= 1)
		i = 63;
	else
		i = find_first_bit(&bitfield, sizeof(unsigned long)*8);
	memnodemapsize = (memtop >> i)+1;
	return i;
}

int __init compute_hash_shift(struct bootnode *nodes, int numnodes,
			      int *nodeids)
{
	int shift;

	shift = extract_lsb_from_nodes(nodes, numnodes);
	if (allocate_cachealigned_memnodemap())
		return -1;
	printk(KERN_DEBUG "NUMA: Using %d for the hash shift.\n",
		shift);

	if (populate_memnodemap(nodes, numnodes, shift, nodeids) != 1) {
		printk(KERN_INFO "Your memory is not aligned you need to "
		       "rebuild your kernel with a bigger NODEMAPSIZE "
		       "shift=%d\n", shift);
		return -1;
	}
	return shift;
}

int early_pfn_to_nid(unsigned long pfn)
{
	return phys_to_nid(pfn << PAGE_SHIFT);
}

static void * __init early_node_mem(int nodeid, unsigned long start,
				    unsigned long end, unsigned long size,
				    unsigned long align)
{
	unsigned long mem = find_e820_area(start, end, size, align);
	void *ptr;

	if (mem != -1L)
		return __va(mem);

	ptr = __alloc_bootmem_nopanic(size, align, __pa(MAX_DMA_ADDRESS));
	if (ptr == NULL) {
		printk(KERN_ERR "Cannot find %lu bytes in node %d\n",
		       size, nodeid);
		return NULL;
	}
	return ptr;
}

/* Initialize bootmem allocator for a node */
void __init setup_node_bootmem(int nodeid, unsigned long start,
			       unsigned long end)
{
	unsigned long start_pfn, last_pfn, bootmap_pages, bootmap_size;
	unsigned long bootmap_start, nodedata_phys;
	void *bootmap;
	const int pgdat_size = roundup(sizeof(pg_data_t), PAGE_SIZE);
	int nid;

	start = roundup(start, ZONE_ALIGN);

	printk(KERN_INFO "Bootmem setup node %d %016lx-%016lx\n", nodeid,
	       start, end);

	start_pfn = start >> PAGE_SHIFT;
	last_pfn = end >> PAGE_SHIFT;

	node_data[nodeid] = early_node_mem(nodeid, start, end, pgdat_size,
					   SMP_CACHE_BYTES);
	if (node_data[nodeid] == NULL)
		return;
	nodedata_phys = __pa(node_data[nodeid]);
	printk(KERN_INFO "  NODE_DATA [%016lx - %016lx]\n", nodedata_phys,
		nodedata_phys + pgdat_size - 1);

	memset(NODE_DATA(nodeid), 0, sizeof(pg_data_t));
	NODE_DATA(nodeid)->bdata = &bootmem_node_data[nodeid];
	NODE_DATA(nodeid)->node_start_pfn = start_pfn;
	NODE_DATA(nodeid)->node_spanned_pages = last_pfn - start_pfn;

	/*
	 * Find a place for the bootmem map
	 * nodedata_phys could be on other nodes by alloc_bootmem,
	 * so need to sure bootmap_start not to be small, otherwise
	 * early_node_mem will get that with find_e820_area instead
	 * of alloc_bootmem, that could clash with reserved range
	 */
	bootmap_pages = bootmem_bootmap_pages(last_pfn - start_pfn);
	nid = phys_to_nid(nodedata_phys);
	if (nid == nodeid)
		bootmap_start = roundup(nodedata_phys + pgdat_size, PAGE_SIZE);
	else
		bootmap_start = roundup(start, PAGE_SIZE);
	/*
	 * SMP_CACHE_BYTES could be enough, but init_bootmem_node like
	 * to use that to align to PAGE_SIZE
	 */
	bootmap = early_node_mem(nodeid, bootmap_start, end,
				 bootmap_pages<<PAGE_SHIFT, PAGE_SIZE);
	if (bootmap == NULL)  {
		if (nodedata_phys < start || nodedata_phys >= end)
			free_bootmem(nodedata_phys, pgdat_size);
		node_data[nodeid] = NULL;
		return;
	}
	bootmap_start = __pa(bootmap);

	bootmap_size = init_bootmem_node(NODE_DATA(nodeid),
					 bootmap_start >> PAGE_SHIFT,
					 start_pfn, last_pfn);

	printk(KERN_INFO "  bootmap [%016lx -  %016lx] pages %lx\n",
		 bootmap_start, bootmap_start + bootmap_size - 1,
		 bootmap_pages);

	free_bootmem_with_active_regions(nodeid, end);

	/*
	 * convert early reserve to bootmem reserve earlier
	 * otherwise early_node_mem could use early reserved mem
	 * on previous node
	 */
	early_res_to_bootmem(start, end);

	/*
	 * in some case early_node_mem could use alloc_bootmem
	 * to get range on other node, don't reserve that again
	 */
	if (nid != nodeid)
		printk(KERN_INFO "    NODE_DATA(%d) on node %d\n", nodeid, nid);
	else
		reserve_bootmem_node(NODE_DATA(nodeid), nodedata_phys,
					pgdat_size, BOOTMEM_DEFAULT);
	nid = phys_to_nid(bootmap_start);
	if (nid != nodeid)
		printk(KERN_INFO "    bootmap(%d) on node %d\n", nodeid, nid);
	else
		reserve_bootmem_node(NODE_DATA(nodeid), bootmap_start,
				 bootmap_pages<<PAGE_SHIFT, BOOTMEM_DEFAULT);

#ifdef CONFIG_ACPI_NUMA
	srat_reserve_add_area(nodeid);
#endif
	node_set_online(nodeid);
}

/*
 * There are unfortunately some poorly designed mainboards around that
 * only connect memory to a single CPU. This breaks the 1:1 cpu->node
 * mapping. To avoid this fill in the mapping for all possible CPUs,
 * as the number of CPUs is not known yet. We round robin the existing
 * nodes.
 */
void __init numa_init_array(void)
{
	int rr, i;

	rr = first_node(node_online_map);
	for (i = 0; i < nr_cpu_ids; i++) {
		if (early_cpu_to_node(i) != NUMA_NO_NODE)
			continue;
		numa_set_node(i, rr);
		rr = next_node(rr, node_online_map);
		if (rr == MAX_NUMNODES)
			rr = first_node(node_online_map);
	}
}

#ifdef CONFIG_NUMA_EMU
/* Numa emulation */
static char *cmdline __initdata;

/*
 * Setups up nid to range from addr to addr + size.  If the end
 * boundary is greater than max_addr, then max_addr is used instead.
 * The return value is 0 if there is additional memory left for
 * allocation past addr and -1 otherwise.  addr is adjusted to be at
 * the end of the node.
 */
static int __init setup_node_range(int nid, struct bootnode *nodes, u64 *addr,
				   u64 size, u64 max_addr)
{
	int ret = 0;

	nodes[nid].start = *addr;
	*addr += size;
	if (*addr >= max_addr) {
		*addr = max_addr;
		ret = -1;
	}
	nodes[nid].end = *addr;
	node_set(nid, node_possible_map);
	printk(KERN_INFO "Faking node %d at %016Lx-%016Lx (%LuMB)\n", nid,
	       nodes[nid].start, nodes[nid].end,
	       (nodes[nid].end - nodes[nid].start) >> 20);
	return ret;
}

/*
 * Splits num_nodes nodes up equally starting at node_start.  The return value
 * is the number of nodes split up and addr is adjusted to be at the end of the
 * last node allocated.
 */
static int __init split_nodes_equally(struct bootnode *nodes, u64 *addr,
				      u64 max_addr, int node_start,
				      int num_nodes)
{
	unsigned int big;
	u64 size;
	int i;

	if (num_nodes <= 0)
		return -1;
	if (num_nodes > MAX_NUMNODES)
		num_nodes = MAX_NUMNODES;
	size = (max_addr - *addr - e820_hole_size(*addr, max_addr)) /
	       num_nodes;
	/*
	 * Calculate the number of big nodes that can be allocated as a result
	 * of consolidating the leftovers.
	 */
	big = ((size & ~FAKE_NODE_MIN_HASH_MASK) * num_nodes) /
	      FAKE_NODE_MIN_SIZE;

	/* Round down to nearest FAKE_NODE_MIN_SIZE. */
	size &= FAKE_NODE_MIN_HASH_MASK;
	if (!size) {
		printk(KERN_ERR "Not enough memory for each node.  "
		       "NUMA emulation disabled.\n");
		return -1;
	}

	for (i = node_start; i < num_nodes + node_start; i++) {
		u64 end = *addr + size;

		if (i < big)
			end += FAKE_NODE_MIN_SIZE;
		/*
		 * The final node can have the remaining system RAM.  Other
		 * nodes receive roughly the same amount of available pages.
		 */
		if (i == num_nodes + node_start - 1)
			end = max_addr;
		else
			while (end - *addr - e820_hole_size(*addr, end) <
			       size) {
				end += FAKE_NODE_MIN_SIZE;
				if (end > max_addr) {
					end = max_addr;
					break;
				}
			}
		if (setup_node_range(i, nodes, addr, end - *addr, max_addr) < 0)
			break;
	}
	return i - node_start + 1;
}

/*
 * Splits the remaining system RAM into chunks of size.  The remaining memory is
 * always assigned to a final node and can be asymmetric.  Returns the number of
 * nodes split.
 */
static int __init split_nodes_by_size(struct bootnode *nodes, u64 *addr,
				      u64 max_addr, int node_start, u64 size)
{
	int i = node_start;
	size = (size << 20) & FAKE_NODE_MIN_HASH_MASK;
	while (!setup_node_range(i++, nodes, addr, size, max_addr))
		;
	return i - node_start;
}

/*
 * Sets up the system RAM area from start_pfn to last_pfn according to the
 * numa=fake command-line option.
 */
static struct bootnode nodes[MAX_NUMNODES] __initdata;

static int __init numa_emulation(unsigned long start_pfn, unsigned long last_pfn)
{
	u64 size, addr = start_pfn << PAGE_SHIFT;
	u64 max_addr = last_pfn << PAGE_SHIFT;
	int num_nodes = 0, num = 0, coeff_flag, coeff = -1, i;

	memset(&nodes, 0, sizeof(nodes));
	/*
	 * If the numa=fake command-line is just a single number N, split the
	 * system RAM into N fake nodes.
	 */
	if (!strchr(cmdline, '*') && !strchr(cmdline, ',')) {
		long n = simple_strtol(cmdline, NULL, 0);

		num_nodes = split_nodes_equally(nodes, &addr, max_addr, 0, n);
		if (num_nodes < 0)
			return num_nodes;
		goto out;
	}

	/* Parse the command line. */
	for (coeff_flag = 0; ; cmdline++) {
		if (*cmdline && isdigit(*cmdline)) {
			num = num * 10 + *cmdline - '0';
			continue;
		}
		if (*cmdline == '*') {
			if (num > 0)
				coeff = num;
			coeff_flag = 1;
		}
		if (!*cmdline || *cmdline == ',') {
			if (!coeff_flag)
				coeff = 1;
			/*
			 * Round down to the nearest FAKE_NODE_MIN_SIZE.
			 * Command-line coefficients are in megabytes.
			 */
			size = ((u64)num << 20) & FAKE_NODE_MIN_HASH_MASK;
			if (size)
				for (i = 0; i < coeff; i++, num_nodes++)
					if (setup_node_range(num_nodes, nodes,
						&addr, size, max_addr) < 0)
						goto done;
			if (!*cmdline)
				break;
			coeff_flag = 0;
			coeff = -1;
		}
		num = 0;
	}
done:
	if (!num_nodes)
		return -1;
	/* Fill remainder of system RAM, if appropriate. */
	if (addr < max_addr) {
		if (coeff_flag && coeff < 0) {
			/* Split remaining nodes into num-sized chunks */
			num_nodes += split_nodes_by_size(nodes, &addr, max_addr,
							 num_nodes, num);
			goto out;
		}
		switch (*(cmdline - 1)) {
		case '*':
			/* Split remaining nodes into coeff chunks */
			if (coeff <= 0)
				break;
			num_nodes += split_nodes_equally(nodes, &addr, max_addr,
							 num_nodes, coeff);
			break;
		case ',':
			/* Do not allocate remaining system RAM */
			break;
		default:
			/* Give one final node */
			setup_node_range(num_nodes, nodes, &addr,
					 max_addr - addr, max_addr);
			num_nodes++;
		}
	}
out:
	memnode_shift = compute_hash_shift(nodes, num_nodes, NULL);
	if (memnode_shift < 0) {
		memnode_shift = 0;
		printk(KERN_ERR "No NUMA hash function found.  NUMA emulation "
		       "disabled.\n");
		return -1;
	}

	/*
	 * We need to vacate all active ranges that may have been registered by
	 * SRAT and set acpi_numa to -1 so that srat_disabled() always returns
	 * true.  NUMA emulation has succeeded so we will not scan ACPI nodes.
	 */
	remove_all_active_ranges();
#ifdef CONFIG_ACPI_NUMA
	acpi_numa = -1;
#endif
	for_each_node_mask(i, node_possible_map) {
		e820_register_active_regions(i, nodes[i].start >> PAGE_SHIFT,
						nodes[i].end >> PAGE_SHIFT);
		setup_node_bootmem(i, nodes[i].start, nodes[i].end);
	}
	acpi_fake_nodes(nodes, num_nodes);
	numa_init_array();
	return 0;
}
#endif /* CONFIG_NUMA_EMU */

void __init initmem_init(unsigned long start_pfn, unsigned long last_pfn)
{
	int i;

	nodes_clear(node_possible_map);
	nodes_clear(node_online_map);

#ifdef CONFIG_NUMA_EMU
	if (cmdline && !numa_emulation(start_pfn, last_pfn))
		return;
	nodes_clear(node_possible_map);
	nodes_clear(node_online_map);
#endif

#ifdef CONFIG_ACPI_NUMA
	if (!numa_off && !acpi_scan_nodes(start_pfn << PAGE_SHIFT,
					  last_pfn << PAGE_SHIFT))
		return;
	nodes_clear(node_possible_map);
	nodes_clear(node_online_map);
#endif

#ifdef CONFIG_K8_NUMA
	if (!numa_off && !k8_scan_nodes(start_pfn<<PAGE_SHIFT,
					last_pfn<<PAGE_SHIFT))
		return;
	nodes_clear(node_possible_map);
	nodes_clear(node_online_map);
#endif
	printk(KERN_INFO "%s\n",
	       numa_off ? "NUMA turned off" : "No NUMA configuration found");

	printk(KERN_INFO "Faking a node at %016lx-%016lx\n",
	       start_pfn << PAGE_SHIFT,
	       last_pfn << PAGE_SHIFT);
	/* setup dummy node covering all memory */
	memnode_shift = 63;
	memnodemap = memnode.embedded_map;
	memnodemap[0] = 0;
	node_set_online(0);
	node_set(0, node_possible_map);
	for (i = 0; i < nr_cpu_ids; i++)
		numa_set_node(i, 0);
	e820_register_active_regions(0, start_pfn, last_pfn);
	setup_node_bootmem(0, start_pfn << PAGE_SHIFT, last_pfn << PAGE_SHIFT);
}

unsigned long __init numa_free_all_bootmem(void)
{
	unsigned long pages = 0;
	int i;

	for_each_online_node(i)
		pages += free_all_bootmem_node(NODE_DATA(i));

	return pages;
}

void __init paging_init(void)
{
	unsigned long max_zone_pfns[MAX_NR_ZONES];

	memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
	max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN;
	max_zone_pfns[ZONE_DMA32] = MAX_DMA32_PFN;
	max_zone_pfns[ZONE_NORMAL] = max_pfn;

	sparse_memory_present_with_active_regions(MAX_NUMNODES);
	sparse_init();

	free_area_init_nodes(max_zone_pfns);
}

static __init int numa_setup(char *opt)
{
	if (!opt)
		return -EINVAL;
	if (!strncmp(opt, "off", 3))
		numa_off = 1;
#ifdef CONFIG_NUMA_EMU
	if (!strncmp(opt, "fake=", 5))
		cmdline = opt + 5;
#endif
#ifdef CONFIG_ACPI_NUMA
	if (!strncmp(opt, "noacpi", 6))
		acpi_numa = -1;
	if (!strncmp(opt, "hotadd=", 7))
		hotadd_percent = simple_strtoul(opt+7, NULL, 10);
#endif
	return 0;
}
early_param("numa", numa_setup);

#ifdef CONFIG_NUMA
/*
 * Setup early cpu_to_node.
 *
 * Populate cpu_to_node[] only if x86_cpu_to_apicid[],
 * and apicid_to_node[] tables have valid entries for a CPU.
 * This means we skip cpu_to_node[] initialisation for NUMA
 * emulation and faking node case (when running a kernel compiled
 * for NUMA on a non NUMA box), which is OK as cpu_to_node[]
 * is already initialized in a round robin manner at numa_init_array,
 * prior to this call, and this initialization is good enough
 * for the fake NUMA cases.
 *
 * Called before the per_cpu areas are setup.
 */
void __init init_cpu_to_node(void)
{
	int cpu;
	u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid);

	BUG_ON(cpu_to_apicid == NULL);

	for_each_possible_cpu(cpu) {
		int node;
		u16 apicid = cpu_to_apicid[cpu];

		if (apicid == BAD_APICID)
			continue;
		node = apicid_to_node[apicid];
		if (node == NUMA_NO_NODE)
			continue;
		if (!node_online(node))
			continue;
		numa_set_node(cpu, node);
	}
}
#endif