aboutsummaryrefslogtreecommitdiffstats
path: root/arch/x86_64/mm/numa.c
blob: 41b8fb069924d9a6b26582688f9421aa44e80d8d (plain) (blame)
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
/* 
 * 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 <asm/e820.h>
#include <asm/proto.h>
#include <asm/dma.h>
#include <asm/numa.h>
#include <asm/acpi.h>

#ifndef Dprintk
#define Dprintk(x...)
#endif

struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
bootmem_data_t plat_node_bdata[MAX_NUMNODES];

struct memnode memnode;

unsigned char cpu_to_node[NR_CPUS] __read_mostly = {
	[0 ... NR_CPUS-1] = NUMA_NO_NODE
};
unsigned char apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
 	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
};
cpumask_t node_to_cpumask[MAX_NUMNODES] __read_mostly;

int numa_off __initdata;
unsigned long __initdata nodemap_addr;
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 i; 
	int res = -1;
	unsigned long addr, end;

	memset(memnodemap, 0xff, 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] != 0xff)
				return -1;
			memnodemap[addr >> shift] = i;
			addr += (1UL << shift);
		} while (addr < end);
		res = 1;
	} 
	return res;
}

static int __init allocate_cachealigned_memnodemap(void)
{
	unsigned long pad, pad_addr;

	memnodemap = memnode.embedded_map;
	if (memnodemapsize <= 48)
		return 0;

	pad = L1_CACHE_BYTES - 1;
	pad_addr = 0x8000;
	nodemap_size = pad + memnodemapsize;
	nodemap_addr = find_e820_area(pad_addr, end_pfn<<PAGE_SHIFT,
				      nodemap_size);
	if (nodemap_addr == -1UL) {
		printk(KERN_ERR
		       "NUMA: Unable to allocate Memory to Node hash map\n");
		nodemap_addr = nodemap_size = 0;
		return -1;
	}
	pad_addr = (nodemap_addr + pad) & ~pad;
	memnodemap = phys_to_virt(pad_addr);

	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 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) != 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;
}

#ifdef CONFIG_SPARSEMEM
int early_pfn_to_nid(unsigned long pfn)
{
	return phys_to_nid(pfn << PAGE_SHIFT);
}
#endif

static void * __init
early_node_mem(int nodeid, unsigned long start, unsigned long end,
	      unsigned long size)
{
	unsigned long mem = find_e820_area(start, end, size);
	void *ptr;
	if (mem != -1L)
		return __va(mem);
	ptr = __alloc_bootmem_nopanic(size,
				SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS));
	if (ptr == 0) {
		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, end_pfn, bootmap_pages, bootmap_size, bootmap_start; 
	unsigned long nodedata_phys;
	void *bootmap;
	const int pgdat_size = round_up(sizeof(pg_data_t), PAGE_SIZE);

	start = round_up(start, ZONE_ALIGN); 

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

	start_pfn = start >> PAGE_SHIFT;
	end_pfn = end >> PAGE_SHIFT;

	node_data[nodeid] = early_node_mem(nodeid, start, end, pgdat_size);
	if (node_data[nodeid] == NULL)
		return;
	nodedata_phys = __pa(node_data[nodeid]);

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

	/* Find a place for the bootmem map */
	bootmap_pages = bootmem_bootmap_pages(end_pfn - start_pfn); 
	bootmap_start = round_up(nodedata_phys + pgdat_size, PAGE_SIZE);
	bootmap = early_node_mem(nodeid, bootmap_start, end,
					bootmap_pages<<PAGE_SHIFT);
	if (bootmap == NULL)  {
		if (nodedata_phys < start || nodedata_phys >= end)
			free_bootmem((unsigned long)node_data[nodeid],pgdat_size);
		node_data[nodeid] = NULL;
		return;
	}
	bootmap_start = __pa(bootmap);
	Dprintk("bootmap start %lu pages %lu\n", bootmap_start, bootmap_pages); 
	
	bootmap_size = init_bootmem_node(NODE_DATA(nodeid),
					 bootmap_start >> PAGE_SHIFT, 
					 start_pfn, end_pfn); 

	free_bootmem_with_active_regions(nodeid, end);

	reserve_bootmem_node(NODE_DATA(nodeid), nodedata_phys, pgdat_size); 
	reserve_bootmem_node(NODE_DATA(nodeid), bootmap_start, bootmap_pages<<PAGE_SHIFT);
#ifdef CONFIG_ACPI_NUMA
	srat_reserve_add_area(nodeid);
#endif
	node_set_online(nodeid);
} 

/* Initialize final allocator for a zone */
void __init setup_node_zones(int nodeid)
{ 
	unsigned long start_pfn, end_pfn, memmapsize, limit;

 	start_pfn = node_start_pfn(nodeid);
 	end_pfn = node_end_pfn(nodeid);

	Dprintk(KERN_INFO "Setting up memmap for node %d %lx-%lx\n",
		nodeid, start_pfn, end_pfn);

	/* Try to allocate mem_map at end to not fill up precious <4GB
	   memory. */
	memmapsize = sizeof(struct page) * (end_pfn-start_pfn);
	limit = end_pfn << PAGE_SHIFT;
#ifdef CONFIG_FLAT_NODE_MEM_MAP
	NODE_DATA(nodeid)->node_mem_map = 
		__alloc_bootmem_core(NODE_DATA(nodeid)->bdata, 
				memmapsize, SMP_CACHE_BYTES, 
				round_down(limit - memmapsize, PAGE_SIZE), 
				limit);
#endif
} 

void __init numa_init_array(void)
{
	int rr, i;
	/* 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. */
	rr = first_node(node_online_map);
	for (i = 0; i < NR_CPUS; i++) {
		if (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 */
int numa_fake __initdata = 0;

/*
 * This function is used to find out if the start and end correspond to
 * different zones.
 */
int zone_cross_over(unsigned long start, unsigned long end)
{
	if ((start < (MAX_DMA32_PFN << PAGE_SHIFT)) &&
			(end >= (MAX_DMA32_PFN << PAGE_SHIFT)))
		return 1;
	return 0;
}

static int __init numa_emulation(unsigned long start_pfn, unsigned long end_pfn)
{
 	int i, big;
 	struct bootnode nodes[MAX_NUMNODES];
 	unsigned long sz, old_sz;
	unsigned long hole_size;
	unsigned long start, end;
	unsigned long max_addr = (end_pfn << PAGE_SHIFT);

	start = (start_pfn << PAGE_SHIFT);
	hole_size = e820_hole_size(start, max_addr);
	sz = (max_addr - start - hole_size) / numa_fake;

 	/* Kludge needed for the hash function */

	old_sz = sz;
	/*
	 * Round down to the nearest FAKE_NODE_MIN_SIZE.
	 */
	sz &= FAKE_NODE_MIN_HASH_MASK;

	/*
	 * We ensure that each node is at least 64MB big.  Smaller than this
	 * size can cause VM hiccups.
	 */
	if (sz == 0) {
		printk(KERN_INFO "Not enough memory for %d nodes.  Reducing "
				"the number of nodes\n", numa_fake);
		numa_fake = (max_addr - start - hole_size) / FAKE_NODE_MIN_SIZE;
		printk(KERN_INFO "Number of fake nodes will be = %d\n",
				numa_fake);
		sz = FAKE_NODE_MIN_SIZE;
	}
	/*
	 * Find out how many nodes can get an extra NODE_MIN_SIZE granule.
	 * This logic ensures the extra memory gets distributed among as many
	 * nodes as possible (as compared to one single node getting all that
	 * extra memory.
	 */
	big = ((old_sz - sz) * numa_fake) / FAKE_NODE_MIN_SIZE;
	printk(KERN_INFO "Fake node Size: %luMB hole_size: %luMB big nodes: "
			"%d\n",
			(sz >> 20), (hole_size >> 20), big);
 	memset(&nodes,0,sizeof(nodes));
	end = start;
 	for (i = 0; i < numa_fake; i++) {
		/*
		 * In case we are not able to allocate enough memory for all
		 * the nodes, we reduce the number of fake nodes.
		 */
		if (end >= max_addr) {
			numa_fake = i - 1;
			break;
		}
 		start = nodes[i].start = end;
		/*
		 * Final node can have all the remaining memory.
		 */
 		if (i == numa_fake-1)
 			sz = max_addr - start;
 		end = nodes[i].start + sz;
		/*
		 * Fir "big" number of nodes get extra granule.
		 */
		if (i < big)
			end += FAKE_NODE_MIN_SIZE;
		/*
		 * Iterate over the range to ensure that this node gets at
		 * least sz amount of RAM (excluding holes)
		 */
		while ((end - start - e820_hole_size(start, end)) < sz) {
			end += FAKE_NODE_MIN_SIZE;
			if (end >= max_addr)
				break;
		}
		/*
		 * Look at the next node to make sure there is some real memory
		 * to map.  Bad things happen when the only memory present
		 * in a zone on a fake node is IO hole.
		 */
		while (e820_hole_size(end, end + FAKE_NODE_MIN_SIZE) > 0) {
			if (zone_cross_over(start, end + sz)) {
				end = (MAX_DMA32_PFN << PAGE_SHIFT);
				break;
			}
			if (end >= max_addr)
				break;
			end += FAKE_NODE_MIN_SIZE;
		}
		if (end > max_addr)
			end = max_addr;
		nodes[i].end = end;
 		printk(KERN_INFO "Faking node %d at %016Lx-%016Lx (%LuMB)\n",
 		       i,
 		       nodes[i].start, nodes[i].end,
 		       (nodes[i].end - nodes[i].start) >> 20);
		node_set_online(i);
 	}
 	memnode_shift = compute_hash_shift(nodes, numa_fake);
 	if (memnode_shift < 0) {
 		memnode_shift = 0;
 		printk(KERN_ERR "No NUMA hash function found. Emulation disabled.\n");
 		return -1;
 	}
 	for_each_online_node(i) {
		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);
	}
 	numa_init_array();
 	return 0;
}
#endif

void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
{ 
	int i;

#ifdef CONFIG_NUMA_EMU
	if (numa_fake && !numa_emulation(start_pfn, end_pfn))
 		return;
#endif

#ifdef CONFIG_ACPI_NUMA
	if (!numa_off && !acpi_scan_nodes(start_pfn << PAGE_SHIFT,
					  end_pfn << PAGE_SHIFT))
 		return;
#endif

#ifdef CONFIG_K8_NUMA
	if (!numa_off && !k8_scan_nodes(start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT))
		return;
#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,
	       end_pfn << PAGE_SHIFT); 
		/* setup dummy node covering all memory */ 
	memnode_shift = 63; 
	memnodemap = memnode.embedded_map;
	memnodemap[0] = 0;
	nodes_clear(node_online_map);
	node_set_online(0);
	for (i = 0; i < NR_CPUS; i++)
		numa_set_node(i, 0);
	node_to_cpumask[0] = cpumask_of_cpu(0);
	e820_register_active_regions(0, start_pfn, end_pfn);
	setup_node_bootmem(0, start_pfn << PAGE_SHIFT, end_pfn << PAGE_SHIFT);
}

__cpuinit void numa_add_cpu(int cpu)
{
	set_bit(cpu, &node_to_cpumask[cpu_to_node(cpu)]);
} 

void __cpuinit numa_set_node(int cpu, int node)
{
	cpu_pda(cpu)->nodenumber = node;
	cpu_to_node[cpu] = node;
}

unsigned long __init numa_free_all_bootmem(void) 
{ 
	int i;
	unsigned long pages = 0;
	for_each_online_node(i) {
		pages += free_all_bootmem_node(NODE_DATA(i));
	}
	return pages;
} 

void __init paging_init(void)
{ 
	int i;
	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] = end_pfn;

	sparse_memory_present_with_active_regions(MAX_NUMNODES);
	sparse_init();

	for_each_online_node(i) {
		setup_node_zones(i); 
	}

	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)) {
		numa_fake = simple_strtoul(opt+5,NULL,0); ;
		if (numa_fake >= MAX_NUMNODES)
			numa_fake = MAX_NUMNODES;
	}
#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);

/*
 * 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.
 */
void __init init_cpu_to_node(void)
{
	int i;
 	for (i = 0; i < NR_CPUS; i++) {
		u8 apicid = x86_cpu_to_apicid[i];
		if (apicid == BAD_APICID)
			continue;
		if (apicid_to_node[apicid] == NUMA_NO_NODE)
			continue;
		numa_set_node(i,apicid_to_node[apicid]);
	}
}

EXPORT_SYMBOL(cpu_to_node);
EXPORT_SYMBOL(node_to_cpumask);
EXPORT_SYMBOL(memnode);
EXPORT_SYMBOL(node_data);

#ifdef CONFIG_DISCONTIGMEM
/*
 * Functions to convert PFNs from/to per node page addresses.
 * These are out of line because they are quite big.
 * They could be all tuned by pre caching more state.
 * Should do that.
 */

int pfn_valid(unsigned long pfn)
{
	unsigned nid;
	if (pfn >= num_physpages)
		return 0;
	nid = pfn_to_nid(pfn);
	if (nid == 0xff)
		return 0;
	return pfn >= node_start_pfn(nid) && (pfn) < node_end_pfn(nid);
}
EXPORT_SYMBOL(pfn_valid);
#endif