summaryrefslogtreecommitdiffstats
path: root/arch/parisc/mm/init.c
blob: b0f96c0e6316f15531afb2a5a2a0e4684d1892a8 (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
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
/*
 *  linux/arch/parisc/mm/init.c
 *
 *  Copyright (C) 1995	Linus Torvalds
 *  Copyright 1999 SuSE GmbH
 *    changed by Philipp Rumpf
 *  Copyright 1999 Philipp Rumpf (prumpf@tux.org)
 *  Copyright 2004 Randolph Chung (tausq@debian.org)
 *  Copyright 2006-2007 Helge Deller (deller@gmx.de)
 *
 */


#include <linux/module.h>
#include <linux/mm.h>
#include <linux/bootmem.h>
#include <linux/gfp.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/pci.h>		/* for hppa_dma_ops and pcxl_dma_ops */
#include <linux/initrd.h>
#include <linux/swap.h>
#include <linux/unistd.h>
#include <linux/nodemask.h>	/* for node_online_map */
#include <linux/pagemap.h>	/* for release_pages and page_cache_release */

#include <asm/pgalloc.h>
#include <asm/pgtable.h>
#include <asm/tlb.h>
#include <asm/pdc_chassis.h>
#include <asm/mmzone.h>
#include <asm/sections.h>

extern int  data_start;

#if PT_NLEVELS == 3
/* NOTE: This layout exactly conforms to the hybrid L2/L3 page table layout
 * with the first pmd adjacent to the pgd and below it. gcc doesn't actually
 * guarantee that global objects will be laid out in memory in the same order
 * as the order of declaration, so put these in different sections and use
 * the linker script to order them. */
pmd_t pmd0[PTRS_PER_PMD] __attribute__ ((__section__ (".data..vm0.pmd"), aligned(PAGE_SIZE)));
#endif

pgd_t swapper_pg_dir[PTRS_PER_PGD] __attribute__ ((__section__ (".data..vm0.pgd"), aligned(PAGE_SIZE)));
pte_t pg0[PT_INITIAL * PTRS_PER_PTE] __attribute__ ((__section__ (".data..vm0.pte"), aligned(PAGE_SIZE)));

#ifdef CONFIG_DISCONTIGMEM
struct node_map_data node_data[MAX_NUMNODES] __read_mostly;
signed char pfnnid_map[PFNNID_MAP_MAX] __read_mostly;
#endif

static struct resource data_resource = {
	.name	= "Kernel data",
	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM,
};

static struct resource code_resource = {
	.name	= "Kernel code",
	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM,
};

static struct resource pdcdata_resource = {
	.name	= "PDC data (Page Zero)",
	.start	= 0,
	.end	= 0x9ff,
	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM,
};

static struct resource sysram_resources[MAX_PHYSMEM_RANGES] __read_mostly;

/* The following array is initialized from the firmware specific
 * information retrieved in kernel/inventory.c.
 */

physmem_range_t pmem_ranges[MAX_PHYSMEM_RANGES] __read_mostly;
int npmem_ranges __read_mostly;

#ifdef CONFIG_64BIT
#define MAX_MEM         (~0UL)
#else /* !CONFIG_64BIT */
#define MAX_MEM         (3584U*1024U*1024U)
#endif /* !CONFIG_64BIT */

static unsigned long mem_limit __read_mostly = MAX_MEM;

static void __init mem_limit_func(void)
{
	char *cp, *end;
	unsigned long limit;

	/* We need this before __setup() functions are called */

	limit = MAX_MEM;
	for (cp = boot_command_line; *cp; ) {
		if (memcmp(cp, "mem=", 4) == 0) {
			cp += 4;
			limit = memparse(cp, &end);
			if (end != cp)
				break;
			cp = end;
		} else {
			while (*cp != ' ' && *cp)
				++cp;
			while (*cp == ' ')
				++cp;
		}
	}

	if (limit < mem_limit)
		mem_limit = limit;
}

#define MAX_GAP (0x40000000UL >> PAGE_SHIFT)

static void __init setup_bootmem(void)
{
	unsigned long bootmap_size;
	unsigned long mem_max;
	unsigned long bootmap_pages;
	unsigned long bootmap_start_pfn;
	unsigned long bootmap_pfn;
#ifndef CONFIG_DISCONTIGMEM
	physmem_range_t pmem_holes[MAX_PHYSMEM_RANGES - 1];
	int npmem_holes;
#endif
	int i, sysram_resource_count;

	disable_sr_hashing(); /* Turn off space register hashing */

	/*
	 * Sort the ranges. Since the number of ranges is typically
	 * small, and performance is not an issue here, just do
	 * a simple insertion sort.
	 */

	for (i = 1; i < npmem_ranges; i++) {
		int j;

		for (j = i; j > 0; j--) {
			unsigned long tmp;

			if (pmem_ranges[j-1].start_pfn <
			    pmem_ranges[j].start_pfn) {

				break;
			}
			tmp = pmem_ranges[j-1].start_pfn;
			pmem_ranges[j-1].start_pfn = pmem_ranges[j].start_pfn;
			pmem_ranges[j].start_pfn = tmp;
			tmp = pmem_ranges[j-1].pages;
			pmem_ranges[j-1].pages = pmem_ranges[j].pages;
			pmem_ranges[j].pages = tmp;
		}
	}

#ifndef CONFIG_DISCONTIGMEM
	/*
	 * Throw out ranges that are too far apart (controlled by
	 * MAX_GAP).
	 */

	for (i = 1; i < npmem_ranges; i++) {
		if (pmem_ranges[i].start_pfn -
			(pmem_ranges[i-1].start_pfn +
			 pmem_ranges[i-1].pages) > MAX_GAP) {
			npmem_ranges = i;
			printk("Large gap in memory detected (%ld pages). "
			       "Consider turning on CONFIG_DISCONTIGMEM\n",
			       pmem_ranges[i].start_pfn -
			       (pmem_ranges[i-1].start_pfn +
			        pmem_ranges[i-1].pages));
			break;
		}
	}
#endif

	if (npmem_ranges > 1) {

		/* Print the memory ranges */

		printk(KERN_INFO "Memory Ranges:\n");

		for (i = 0; i < npmem_ranges; i++) {
			unsigned long start;
			unsigned long size;

			size = (pmem_ranges[i].pages << PAGE_SHIFT);
			start = (pmem_ranges[i].start_pfn << PAGE_SHIFT);
			printk(KERN_INFO "%2d) Start 0x%016lx End 0x%016lx Size %6ld MB\n",
				i,start, start + (size - 1), size >> 20);
		}
	}

	sysram_resource_count = npmem_ranges;
	for (i = 0; i < sysram_resource_count; i++) {
		struct resource *res = &sysram_resources[i];
		res->name = "System RAM";
		res->start = pmem_ranges[i].start_pfn << PAGE_SHIFT;
		res->end = res->start + (pmem_ranges[i].pages << PAGE_SHIFT)-1;
		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
		request_resource(&iomem_resource, res);
	}

	/*
	 * For 32 bit kernels we limit the amount of memory we can
	 * support, in order to preserve enough kernel address space
	 * for other purposes. For 64 bit kernels we don't normally
	 * limit the memory, but this mechanism can be used to
	 * artificially limit the amount of memory (and it is written
	 * to work with multiple memory ranges).
	 */

	mem_limit_func();       /* check for "mem=" argument */

	mem_max = 0;
	for (i = 0; i < npmem_ranges; i++) {
		unsigned long rsize;

		rsize = pmem_ranges[i].pages << PAGE_SHIFT;
		if ((mem_max + rsize) > mem_limit) {
			printk(KERN_WARNING "Memory truncated to %ld MB\n", mem_limit >> 20);
			if (mem_max == mem_limit)
				npmem_ranges = i;
			else {
				pmem_ranges[i].pages =   (mem_limit >> PAGE_SHIFT)
						       - (mem_max >> PAGE_SHIFT);
				npmem_ranges = i + 1;
				mem_max = mem_limit;
			}
			break;
		}
		mem_max += rsize;
	}

	printk(KERN_INFO "Total Memory: %ld MB\n",mem_max >> 20);

#ifndef CONFIG_DISCONTIGMEM
	/* Merge the ranges, keeping track of the holes */

	{
		unsigned long end_pfn;
		unsigned long hole_pages;

		npmem_holes = 0;
		end_pfn = pmem_ranges[0].start_pfn + pmem_ranges[0].pages;
		for (i = 1; i < npmem_ranges; i++) {

			hole_pages = pmem_ranges[i].start_pfn - end_pfn;
			if (hole_pages) {
				pmem_holes[npmem_holes].start_pfn = end_pfn;
				pmem_holes[npmem_holes++].pages = hole_pages;
				end_pfn += hole_pages;
			}
			end_pfn += pmem_ranges[i].pages;
		}

		pmem_ranges[0].pages = end_pfn - pmem_ranges[0].start_pfn;
		npmem_ranges = 1;
	}
#endif

	bootmap_pages = 0;
	for (i = 0; i < npmem_ranges; i++)
		bootmap_pages += bootmem_bootmap_pages(pmem_ranges[i].pages);

	bootmap_start_pfn = PAGE_ALIGN(__pa((unsigned long) &_end)) >> PAGE_SHIFT;

#ifdef CONFIG_DISCONTIGMEM
	for (i = 0; i < MAX_PHYSMEM_RANGES; i++) {
		memset(NODE_DATA(i), 0, sizeof(pg_data_t));
		NODE_DATA(i)->bdata = &bootmem_node_data[i];
	}
	memset(pfnnid_map, 0xff, sizeof(pfnnid_map));

	for (i = 0; i < npmem_ranges; i++) {
		node_set_state(i, N_NORMAL_MEMORY);
		node_set_online(i);
	}
#endif

	/*
	 * Initialize and free the full range of memory in each range.
	 * Note that the only writing these routines do are to the bootmap,
	 * and we've made sure to locate the bootmap properly so that they
	 * won't be writing over anything important.
	 */

	bootmap_pfn = bootmap_start_pfn;
	max_pfn = 0;
	for (i = 0; i < npmem_ranges; i++) {
		unsigned long start_pfn;
		unsigned long npages;

		start_pfn = pmem_ranges[i].start_pfn;
		npages = pmem_ranges[i].pages;

		bootmap_size = init_bootmem_node(NODE_DATA(i),
						bootmap_pfn,
						start_pfn,
						(start_pfn + npages) );
		free_bootmem_node(NODE_DATA(i),
				  (start_pfn << PAGE_SHIFT),
				  (npages << PAGE_SHIFT) );
		bootmap_pfn += (bootmap_size + PAGE_SIZE - 1) >> PAGE_SHIFT;
		if ((start_pfn + npages) > max_pfn)
			max_pfn = start_pfn + npages;
	}

	/* IOMMU is always used to access "high mem" on those boxes
	 * that can support enough mem that a PCI device couldn't
	 * directly DMA to any physical addresses.
	 * ISA DMA support will need to revisit this.
	 */
	max_low_pfn = max_pfn;

	/* bootmap sizing messed up? */
	BUG_ON((bootmap_pfn - bootmap_start_pfn) != bootmap_pages);

	/* reserve PAGE0 pdc memory, kernel text/data/bss & bootmap */

#define PDC_CONSOLE_IO_IODC_SIZE 32768

	reserve_bootmem_node(NODE_DATA(0), 0UL,
			(unsigned long)(PAGE0->mem_free +
				PDC_CONSOLE_IO_IODC_SIZE), BOOTMEM_DEFAULT);
	reserve_bootmem_node(NODE_DATA(0), __pa((unsigned long)_text),
			(unsigned long)(_end - _text), BOOTMEM_DEFAULT);
	reserve_bootmem_node(NODE_DATA(0), (bootmap_start_pfn << PAGE_SHIFT),
			((bootmap_pfn - bootmap_start_pfn) << PAGE_SHIFT),
			BOOTMEM_DEFAULT);

#ifndef CONFIG_DISCONTIGMEM

	/* reserve the holes */

	for (i = 0; i < npmem_holes; i++) {
		reserve_bootmem_node(NODE_DATA(0),
				(pmem_holes[i].start_pfn << PAGE_SHIFT),
				(pmem_holes[i].pages << PAGE_SHIFT),
				BOOTMEM_DEFAULT);
	}
#endif

#ifdef CONFIG_BLK_DEV_INITRD
	if (initrd_start) {
		printk(KERN_INFO "initrd: %08lx-%08lx\n", initrd_start, initrd_end);
		if (__pa(initrd_start) < mem_max) {
			unsigned long initrd_reserve;

			if (__pa(initrd_end) > mem_max) {
				initrd_reserve = mem_max - __pa(initrd_start);
			} else {
				initrd_reserve = initrd_end - initrd_start;
			}
			initrd_below_start_ok = 1;
			printk(KERN_INFO "initrd: reserving %08lx-%08lx (mem_max %08lx)\n", __pa(initrd_start), __pa(initrd_start) + initrd_reserve, mem_max);

			reserve_bootmem_node(NODE_DATA(0), __pa(initrd_start),
					initrd_reserve, BOOTMEM_DEFAULT);
		}
	}
#endif

	data_resource.start =  virt_to_phys(&data_start);
	data_resource.end = virt_to_phys(_end) - 1;
	code_resource.start = virt_to_phys(_text);
	code_resource.end = virt_to_phys(&data_start)-1;

	/* We don't know which region the kernel will be in, so try
	 * all of them.
	 */
	for (i = 0; i < sysram_resource_count; i++) {
		struct resource *res = &sysram_resources[i];
		request_resource(res, &code_resource);
		request_resource(res, &data_resource);
	}
	request_resource(&sysram_resources[0], &pdcdata_resource);
}

static void __init map_pages(unsigned long start_vaddr,
			     unsigned long start_paddr, unsigned long size,
			     pgprot_t pgprot, int force)
{
	pgd_t *pg_dir;
	pmd_t *pmd;
	pte_t *pg_table;
	unsigned long end_paddr;
	unsigned long start_pmd;
	unsigned long start_pte;
	unsigned long tmp1;
	unsigned long tmp2;
	unsigned long address;
	unsigned long vaddr;
	unsigned long ro_start;
	unsigned long ro_end;
	unsigned long fv_addr;
	unsigned long gw_addr;
	extern const unsigned long fault_vector_20;
	extern void * const linux_gateway_page;

	ro_start = __pa((unsigned long)_text);
	ro_end   = __pa((unsigned long)&data_start);
	fv_addr  = __pa((unsigned long)&fault_vector_20) & PAGE_MASK;
	gw_addr  = __pa((unsigned long)&linux_gateway_page) & PAGE_MASK;

	end_paddr = start_paddr + size;

	pg_dir = pgd_offset_k(start_vaddr);

#if PTRS_PER_PMD == 1
	start_pmd = 0;
#else
	start_pmd = ((start_vaddr >> PMD_SHIFT) & (PTRS_PER_PMD - 1));
#endif
	start_pte = ((start_vaddr >> PAGE_SHIFT) & (PTRS_PER_PTE - 1));

	address = start_paddr;
	vaddr = start_vaddr;
	while (address < end_paddr) {
#if PTRS_PER_PMD == 1
		pmd = (pmd_t *)__pa(pg_dir);
#else
		pmd = (pmd_t *)pgd_address(*pg_dir);

		/*
		 * pmd is physical at this point
		 */

		if (!pmd) {
			pmd = (pmd_t *) alloc_bootmem_low_pages_node(NODE_DATA(0), PAGE_SIZE << PMD_ORDER);
			pmd = (pmd_t *) __pa(pmd);
		}

		pgd_populate(NULL, pg_dir, __va(pmd));
#endif
		pg_dir++;

		/* now change pmd to kernel virtual addresses */

		pmd = (pmd_t *)__va(pmd) + start_pmd;
		for (tmp1 = start_pmd; tmp1 < PTRS_PER_PMD; tmp1++, pmd++) {

			/*
			 * pg_table is physical at this point
			 */

			pg_table = (pte_t *)pmd_address(*pmd);
			if (!pg_table) {
				pg_table = (pte_t *)
					alloc_bootmem_low_pages_node(NODE_DATA(0), PAGE_SIZE);
				pg_table = (pte_t *) __pa(pg_table);
			}

			pmd_populate_kernel(NULL, pmd, __va(pg_table));

			/* now change pg_table to kernel virtual addresses */

			pg_table = (pte_t *) __va(pg_table) + start_pte;
			for (tmp2 = start_pte; tmp2 < PTRS_PER_PTE; tmp2++, pg_table++) {
				pte_t pte;

				/*
				 * Map the fault vector writable so we can
				 * write the HPMC checksum.
				 */
				if (force)
					pte =  __mk_pte(address, pgprot);
				else if (core_kernel_text(vaddr) &&
					 address != fv_addr)
					pte = __mk_pte(address, PAGE_KERNEL_EXEC);
				else
#if defined(CONFIG_PARISC_PAGE_SIZE_4KB)
				if (address >= ro_start && address < ro_end
							&& address != fv_addr
							&& address != gw_addr)
					pte = __mk_pte(address, PAGE_KERNEL_RO);
				else
#endif
					pte = __mk_pte(address, pgprot);

				if (address >= end_paddr) {
					if (force)
						break;
					else
						pte_val(pte) = 0;
				}

				set_pte(pg_table, pte);

				address += PAGE_SIZE;
				vaddr += PAGE_SIZE;
			}
			start_pte = 0;

			if (address >= end_paddr)
			    break;
		}
		start_pmd = 0;
	}
}

void free_initmem(void)
{
	unsigned long init_begin = (unsigned long)__init_begin;
	unsigned long init_end = (unsigned long)__init_end;

	/* The init text pages are marked R-X.  We have to
	 * flush the icache and mark them RW-
	 *
	 * This is tricky, because map_pages is in the init section.
	 * Do a dummy remap of the data section first (the data
	 * section is already PAGE_KERNEL) to pull in the TLB entries
	 * for map_kernel */
	map_pages(init_begin, __pa(init_begin), init_end - init_begin,
		  PAGE_KERNEL_RWX, 1);
	/* now remap at PAGE_KERNEL since the TLB is pre-primed to execute
	 * map_pages */
	map_pages(init_begin, __pa(init_begin), init_end - init_begin,
		  PAGE_KERNEL, 1);

	/* force the kernel to see the new TLB entries */
	__flush_tlb_range(0, init_begin, init_end);
	/* Attempt to catch anyone trying to execute code here
	 * by filling the page with BRK insns.
	 */
	memset((void *)init_begin, 0x00, init_end - init_begin);
	/* finally dump all the instructions which were cached, since the
	 * pages are no-longer executable */
	flush_icache_range(init_begin, init_end);
	
	free_initmem_default(-1);

	/* set up a new led state on systems shipped LED State panel */
	pdc_chassis_send_status(PDC_CHASSIS_DIRECT_BCOMPLETE);
}


#ifdef CONFIG_DEBUG_RODATA
void mark_rodata_ro(void)
{
	/* rodata memory was already mapped with KERNEL_RO access rights by
           pagetable_init() and map_pages(). No need to do additional stuff here */
	printk (KERN_INFO "Write protecting the kernel read-only data: %luk\n",
		(unsigned long)(__end_rodata - __start_rodata) >> 10);
}
#endif


/*
 * Just an arbitrary offset to serve as a "hole" between mapping areas
 * (between top of physical memory and a potential pcxl dma mapping
 * area, and below the vmalloc mapping area).
 *
 * The current 32K value just means that there will be a 32K "hole"
 * between mapping areas. That means that  any out-of-bounds memory
 * accesses will hopefully be caught. The vmalloc() routines leaves
 * a hole of 4kB between each vmalloced area for the same reason.
 */

 /* Leave room for gateway page expansion */
#if KERNEL_MAP_START < GATEWAY_PAGE_SIZE
#error KERNEL_MAP_START is in gateway reserved region
#endif
#define MAP_START (KERNEL_MAP_START)

#define VM_MAP_OFFSET  (32*1024)
#define SET_MAP_OFFSET(x) ((void *)(((unsigned long)(x) + VM_MAP_OFFSET) \
				     & ~(VM_MAP_OFFSET-1)))

void *parisc_vmalloc_start __read_mostly;
EXPORT_SYMBOL(parisc_vmalloc_start);

#ifdef CONFIG_PA11
unsigned long pcxl_dma_start __read_mostly;
#endif

void __init mem_init(void)
{
	/* Do sanity checks on page table constants */
	BUILD_BUG_ON(PTE_ENTRY_SIZE != sizeof(pte_t));
	BUILD_BUG_ON(PMD_ENTRY_SIZE != sizeof(pmd_t));
	BUILD_BUG_ON(PGD_ENTRY_SIZE != sizeof(pgd_t));
	BUILD_BUG_ON(PAGE_SHIFT + BITS_PER_PTE + BITS_PER_PMD + BITS_PER_PGD
			> BITS_PER_LONG);

	high_memory = __va((max_pfn << PAGE_SHIFT));
	set_max_mapnr(page_to_pfn(virt_to_page(high_memory - 1)) + 1);
	free_all_bootmem();

#ifdef CONFIG_PA11
	if (hppa_dma_ops == &pcxl_dma_ops) {
		pcxl_dma_start = (unsigned long)SET_MAP_OFFSET(MAP_START);
		parisc_vmalloc_start = SET_MAP_OFFSET(pcxl_dma_start
						+ PCXL_DMA_MAP_SIZE);
	} else {
		pcxl_dma_start = 0;
		parisc_vmalloc_start = SET_MAP_OFFSET(MAP_START);
	}
#else
	parisc_vmalloc_start = SET_MAP_OFFSET(MAP_START);
#endif

	mem_init_print_info(NULL);
#ifdef CONFIG_DEBUG_KERNEL /* double-sanity-check paranoia */
	printk("virtual kernel memory layout:\n"
	       "    vmalloc : 0x%p - 0x%p   (%4ld MB)\n"
	       "    memory  : 0x%p - 0x%p   (%4ld MB)\n"
	       "      .init : 0x%p - 0x%p   (%4ld kB)\n"
	       "      .data : 0x%p - 0x%p   (%4ld kB)\n"
	       "      .text : 0x%p - 0x%p   (%4ld kB)\n",

	       (void*)VMALLOC_START, (void*)VMALLOC_END,
	       (VMALLOC_END - VMALLOC_START) >> 20,

	       __va(0), high_memory,
	       ((unsigned long)high_memory - (unsigned long)__va(0)) >> 20,

	       __init_begin, __init_end,
	       ((unsigned long)__init_end - (unsigned long)__init_begin) >> 10,

	       _etext, _edata,
	       ((unsigned long)_edata - (unsigned long)_etext) >> 10,

	       _text, _etext,
	       ((unsigned long)_etext - (unsigned long)_text) >> 10);
#endif
}

unsigned long *empty_zero_page __read_mostly;
EXPORT_SYMBOL(empty_zero_page);

void show_mem(unsigned int filter)
{
	int i,free = 0,total = 0,reserved = 0;
	int shared = 0, cached = 0;

	printk(KERN_INFO "Mem-info:\n");
	show_free_areas(filter);
	if (filter & SHOW_MEM_FILTER_PAGE_COUNT)
		return;
#ifndef CONFIG_DISCONTIGMEM
	i = max_mapnr;
	while (i-- > 0) {
		total++;
		if (PageReserved(mem_map+i))
			reserved++;
		else if (PageSwapCache(mem_map+i))
			cached++;
		else if (!page_count(&mem_map[i]))
			free++;
		else
			shared += page_count(&mem_map[i]) - 1;
	}
#else
	for (i = 0; i < npmem_ranges; i++) {
		int j;

		for (j = node_start_pfn(i); j < node_end_pfn(i); j++) {
			struct page *p;
			unsigned long flags;

			pgdat_resize_lock(NODE_DATA(i), &flags);
			p = nid_page_nr(i, j) - node_start_pfn(i);

			total++;
			if (PageReserved(p))
				reserved++;
			else if (PageSwapCache(p))
				cached++;
			else if (!page_count(p))
				free++;
			else
				shared += page_count(p) - 1;
			pgdat_resize_unlock(NODE_DATA(i), &flags);
        	}
	}
#endif
	printk(KERN_INFO "%d pages of RAM\n", total);
	printk(KERN_INFO "%d reserved pages\n", reserved);
	printk(KERN_INFO "%d pages shared\n", shared);
	printk(KERN_INFO "%d pages swap cached\n", cached);


#ifdef CONFIG_DISCONTIGMEM
	{
		struct zonelist *zl;
		int i, j;

		for (i = 0; i < npmem_ranges; i++) {
			zl = node_zonelist(i, 0);
			for (j = 0; j < MAX_NR_ZONES; j++) {
				struct zoneref *z;
				struct zone *zone;

				printk("Zone list for zone %d on node %d: ", j, i);
				for_each_zone_zonelist(zone, z, zl, j)
					printk("[%d/%s] ", zone_to_nid(zone),
								zone->name);
				printk("\n");
			}
		}
	}
#endif
}

/*
 * pagetable_init() sets up the page tables
 *
 * Note that gateway_init() places the Linux gateway page at page 0.
 * Since gateway pages cannot be dereferenced this has the desirable
 * side effect of trapping those pesky NULL-reference errors in the
 * kernel.
 */
static void __init pagetable_init(void)
{
	int range;

	/* Map each physical memory range to its kernel vaddr */

	for (range = 0; range < npmem_ranges; range++) {
		unsigned long start_paddr;
		unsigned long end_paddr;
		unsigned long size;

		start_paddr = pmem_ranges[range].start_pfn << PAGE_SHIFT;
		end_paddr = start_paddr + (pmem_ranges[range].pages << PAGE_SHIFT);
		size = pmem_ranges[range].pages << PAGE_SHIFT;

		map_pages((unsigned long)__va(start_paddr), start_paddr,
			  size, PAGE_KERNEL, 0);
	}

#ifdef CONFIG_BLK_DEV_INITRD
	if (initrd_end && initrd_end > mem_limit) {
		printk(KERN_INFO "initrd: mapping %08lx-%08lx\n", initrd_start, initrd_end);
		map_pages(initrd_start, __pa(initrd_start),
			  initrd_end - initrd_start, PAGE_KERNEL, 0);
	}
#endif

	empty_zero_page = alloc_bootmem_pages(PAGE_SIZE);
	memset(empty_zero_page, 0, PAGE_SIZE);
}

static void __init gateway_init(void)
{
	unsigned long linux_gateway_page_addr;
	/* FIXME: This is 'const' in order to trick the compiler
	   into not treating it as DP-relative data. */
	extern void * const linux_gateway_page;

	linux_gateway_page_addr = LINUX_GATEWAY_ADDR & PAGE_MASK;

	/*
	 * Setup Linux Gateway page.
	 *
	 * The Linux gateway page will reside in kernel space (on virtual
	 * page 0), so it doesn't need to be aliased into user space.
	 */

	map_pages(linux_gateway_page_addr, __pa(&linux_gateway_page),
		  PAGE_SIZE, PAGE_GATEWAY, 1);
}

#ifdef CONFIG_HPUX
void
map_hpux_gateway_page(struct task_struct *tsk, struct mm_struct *mm)
{
	pgd_t *pg_dir;
	pmd_t *pmd;
	pte_t *pg_table;
	unsigned long start_pmd;
	unsigned long start_pte;
	unsigned long address;
	unsigned long hpux_gw_page_addr;
	/* FIXME: This is 'const' in order to trick the compiler
	   into not treating it as DP-relative data. */
	extern void * const hpux_gateway_page;

	hpux_gw_page_addr = HPUX_GATEWAY_ADDR & PAGE_MASK;

	/*
	 * Setup HP-UX Gateway page.
	 *
	 * The HP-UX gateway page resides in the user address space,
	 * so it needs to be aliased into each process.
	 */

	pg_dir = pgd_offset(mm,hpux_gw_page_addr);

#if PTRS_PER_PMD == 1
	start_pmd = 0;
#else
	start_pmd = ((hpux_gw_page_addr >> PMD_SHIFT) & (PTRS_PER_PMD - 1));
#endif
	start_pte = ((hpux_gw_page_addr >> PAGE_SHIFT) & (PTRS_PER_PTE - 1));

	address = __pa(&hpux_gateway_page);
#if PTRS_PER_PMD == 1
	pmd = (pmd_t *)__pa(pg_dir);
#else
	pmd = (pmd_t *) pgd_address(*pg_dir);

	/*
	 * pmd is physical at this point
	 */

	if (!pmd) {
		pmd = (pmd_t *) get_zeroed_page(GFP_KERNEL);
		pmd = (pmd_t *) __pa(pmd);
	}

	__pgd_val_set(*pg_dir, PxD_FLAG_PRESENT | PxD_FLAG_VALID | (unsigned long) pmd);
#endif
	/* now change pmd to kernel virtual addresses */

	pmd = (pmd_t *)__va(pmd) + start_pmd;

	/*
	 * pg_table is physical at this point
	 */

	pg_table = (pte_t *) pmd_address(*pmd);
	if (!pg_table)
		pg_table = (pte_t *) __pa(get_zeroed_page(GFP_KERNEL));

	__pmd_val_set(*pmd, PxD_FLAG_PRESENT | PxD_FLAG_VALID | (unsigned long) pg_table);

	/* now change pg_table to kernel virtual addresses */

	pg_table = (pte_t *) __va(pg_table) + start_pte;
	set_pte(pg_table, __mk_pte(address, PAGE_GATEWAY));
}
EXPORT_SYMBOL(map_hpux_gateway_page);
#endif

void __init paging_init(void)
{
	int i;

	setup_bootmem();
	pagetable_init();
	gateway_init();
	flush_cache_all_local(); /* start with known state */
	flush_tlb_all_local(NULL);

	for (i = 0; i < npmem_ranges; i++) {
		unsigned long zones_size[MAX_NR_ZONES] = { 0, };

		zones_size[ZONE_NORMAL] = pmem_ranges[i].pages;

#ifdef CONFIG_DISCONTIGMEM
		/* Need to initialize the pfnnid_map before we can initialize
		   the zone */
		{
		    int j;
		    for (j = (pmem_ranges[i].start_pfn >> PFNNID_SHIFT);
			 j <= ((pmem_ranges[i].start_pfn + pmem_ranges[i].pages) >> PFNNID_SHIFT);
			 j++) {
			pfnnid_map[j] = i;
		    }
		}
#endif

		free_area_init_node(i, zones_size,
				pmem_ranges[i].start_pfn, NULL);
	}
}

#ifdef CONFIG_PA20

/*
 * Currently, all PA20 chips have 18 bit protection IDs, which is the
 * limiting factor (space ids are 32 bits).
 */

#define NR_SPACE_IDS 262144

#else

/*
 * Currently we have a one-to-one relationship between space IDs and
 * protection IDs. Older parisc chips (PCXS, PCXT, PCXL, PCXL2) only
 * support 15 bit protection IDs, so that is the limiting factor.
 * PCXT' has 18 bit protection IDs, but only 16 bit spaceids, so it's
 * probably not worth the effort for a special case here.
 */

#define NR_SPACE_IDS 32768

#endif  /* !CONFIG_PA20 */

#define RECYCLE_THRESHOLD (NR_SPACE_IDS / 2)
#define SID_ARRAY_SIZE  (NR_SPACE_IDS / (8 * sizeof(long)))

static unsigned long space_id[SID_ARRAY_SIZE] = { 1 }; /* disallow space 0 */
static unsigned long dirty_space_id[SID_ARRAY_SIZE];
static unsigned long space_id_index;
static unsigned long free_space_ids = NR_SPACE_IDS - 1;
static unsigned long dirty_space_ids = 0;

static DEFINE_SPINLOCK(sid_lock);

unsigned long alloc_sid(void)
{
	unsigned long index;

	spin_lock(&sid_lock);

	if (free_space_ids == 0) {
		if (dirty_space_ids != 0) {
			spin_unlock(&sid_lock);
			flush_tlb_all(); /* flush_tlb_all() calls recycle_sids() */
			spin_lock(&sid_lock);
		}
		BUG_ON(free_space_ids == 0);
	}

	free_space_ids--;

	index = find_next_zero_bit(space_id, NR_SPACE_IDS, space_id_index);
	space_id[index >> SHIFT_PER_LONG] |= (1L << (index & (BITS_PER_LONG - 1)));
	space_id_index = index;

	spin_unlock(&sid_lock);

	return index << SPACEID_SHIFT;
}

void free_sid(unsigned long spaceid)
{
	unsigned long index = spaceid >> SPACEID_SHIFT;
	unsigned long *dirty_space_offset;

	dirty_space_offset = dirty_space_id + (index >> SHIFT_PER_LONG);
	index &= (BITS_PER_LONG - 1);

	spin_lock(&sid_lock);

	BUG_ON(*dirty_space_offset & (1L << index)); /* attempt to free space id twice */

	*dirty_space_offset |= (1L << index);
	dirty_space_ids++;

	spin_unlock(&sid_lock);
}


#ifdef CONFIG_SMP
static void get_dirty_sids(unsigned long *ndirtyptr,unsigned long *dirty_array)
{
	int i;

	/* NOTE: sid_lock must be held upon entry */

	*ndirtyptr = dirty_space_ids;
	if (dirty_space_ids != 0) {
	    for (i = 0; i < SID_ARRAY_SIZE; i++) {
		dirty_array[i] = dirty_space_id[i];
		dirty_space_id[i] = 0;
	    }
	    dirty_space_ids = 0;
	}

	return;
}

static void recycle_sids(unsigned long ndirty,unsigned long *dirty_array)
{
	int i;

	/* NOTE: sid_lock must be held upon entry */

	if (ndirty != 0) {
		for (i = 0; i < SID_ARRAY_SIZE; i++) {
			space_id[i] ^= dirty_array[i];
		}

		free_space_ids += ndirty;
		space_id_index = 0;
	}
}

#else /* CONFIG_SMP */

static void recycle_sids(void)
{
	int i;

	/* NOTE: sid_lock must be held upon entry */

	if (dirty_space_ids != 0) {
		for (i = 0; i < SID_ARRAY_SIZE; i++) {
			space_id[i] ^= dirty_space_id[i];
			dirty_space_id[i] = 0;
		}

		free_space_ids += dirty_space_ids;
		dirty_space_ids = 0;
		space_id_index = 0;
	}
}
#endif

/*
 * flush_tlb_all() calls recycle_sids(), since whenever the entire tlb is
 * purged, we can safely reuse the space ids that were released but
 * not flushed from the tlb.
 */

#ifdef CONFIG_SMP

static unsigned long recycle_ndirty;
static unsigned long recycle_dirty_array[SID_ARRAY_SIZE];
static unsigned int recycle_inuse;

void flush_tlb_all(void)
{
	int do_recycle;

	__inc_irq_stat(irq_tlb_count);
	do_recycle = 0;
	spin_lock(&sid_lock);
	if (dirty_space_ids > RECYCLE_THRESHOLD) {
	    BUG_ON(recycle_inuse);  /* FIXME: Use a semaphore/wait queue here */
	    get_dirty_sids(&recycle_ndirty,recycle_dirty_array);
	    recycle_inuse++;
	    do_recycle++;
	}
	spin_unlock(&sid_lock);
	on_each_cpu(flush_tlb_all_local, NULL, 1);
	if (do_recycle) {
	    spin_lock(&sid_lock);
	    recycle_sids(recycle_ndirty,recycle_dirty_array);
	    recycle_inuse = 0;
	    spin_unlock(&sid_lock);
	}
}
#else
void flush_tlb_all(void)
{
	__inc_irq_stat(irq_tlb_count);
	spin_lock(&sid_lock);
	flush_tlb_all_local(NULL);
	recycle_sids();
	spin_unlock(&sid_lock);
}
#endif

#ifdef CONFIG_BLK_DEV_INITRD
void free_initrd_mem(unsigned long start, unsigned long end)
{
	free_reserved_area((void *)start, (void *)end, -1, "initrd");
}
#endif