aboutsummaryrefslogtreecommitdiffstats
path: root/arch/s390/mm/pgtable.c
blob: c8188a18af05b565e57304b02e153ed4629940d5 (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
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
/*
 *    Copyright IBM Corp. 2007, 2011
 *    Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
 */

#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/gfp.h>
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/smp.h>
#include <linux/highmem.h>
#include <linux/pagemap.h>
#include <linux/spinlock.h>
#include <linux/module.h>
#include <linux/quicklist.h>
#include <linux/rcupdate.h>
#include <linux/slab.h>

#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/tlb.h>
#include <asm/tlbflush.h>
#include <asm/mmu_context.h>

#ifndef CONFIG_64BIT
#define ALLOC_ORDER	1
#define FRAG_MASK	0x0f
#else
#define ALLOC_ORDER	2
#define FRAG_MASK	0x03
#endif


unsigned long *crst_table_alloc(struct mm_struct *mm)
{
	struct page *page = alloc_pages(GFP_KERNEL, ALLOC_ORDER);

	if (!page)
		return NULL;
	return (unsigned long *) page_to_phys(page);
}

void crst_table_free(struct mm_struct *mm, unsigned long *table)
{
	free_pages((unsigned long) table, ALLOC_ORDER);
}

#ifdef CONFIG_64BIT
int crst_table_upgrade(struct mm_struct *mm, unsigned long limit)
{
	unsigned long *table, *pgd;
	unsigned long entry;

	BUG_ON(limit > (1UL << 53));
repeat:
	table = crst_table_alloc(mm);
	if (!table)
		return -ENOMEM;
	spin_lock_bh(&mm->page_table_lock);
	if (mm->context.asce_limit < limit) {
		pgd = (unsigned long *) mm->pgd;
		if (mm->context.asce_limit <= (1UL << 31)) {
			entry = _REGION3_ENTRY_EMPTY;
			mm->context.asce_limit = 1UL << 42;
			mm->context.asce_bits = _ASCE_TABLE_LENGTH |
						_ASCE_USER_BITS |
						_ASCE_TYPE_REGION3;
		} else {
			entry = _REGION2_ENTRY_EMPTY;
			mm->context.asce_limit = 1UL << 53;
			mm->context.asce_bits = _ASCE_TABLE_LENGTH |
						_ASCE_USER_BITS |
						_ASCE_TYPE_REGION2;
		}
		crst_table_init(table, entry);
		pgd_populate(mm, (pgd_t *) table, (pud_t *) pgd);
		mm->pgd = (pgd_t *) table;
		mm->task_size = mm->context.asce_limit;
		table = NULL;
	}
	spin_unlock_bh(&mm->page_table_lock);
	if (table)
		crst_table_free(mm, table);
	if (mm->context.asce_limit < limit)
		goto repeat;
	return 0;
}

void crst_table_downgrade(struct mm_struct *mm, unsigned long limit)
{
	pgd_t *pgd;

	while (mm->context.asce_limit > limit) {
		pgd = mm->pgd;
		switch (pgd_val(*pgd) & _REGION_ENTRY_TYPE_MASK) {
		case _REGION_ENTRY_TYPE_R2:
			mm->context.asce_limit = 1UL << 42;
			mm->context.asce_bits = _ASCE_TABLE_LENGTH |
						_ASCE_USER_BITS |
						_ASCE_TYPE_REGION3;
			break;
		case _REGION_ENTRY_TYPE_R3:
			mm->context.asce_limit = 1UL << 31;
			mm->context.asce_bits = _ASCE_TABLE_LENGTH |
						_ASCE_USER_BITS |
						_ASCE_TYPE_SEGMENT;
			break;
		default:
			BUG();
		}
		mm->pgd = (pgd_t *) (pgd_val(*pgd) & _REGION_ENTRY_ORIGIN);
		mm->task_size = mm->context.asce_limit;
		crst_table_free(mm, (unsigned long *) pgd);
	}
}
#endif

#ifdef CONFIG_PGSTE

/**
 * gmap_alloc - allocate a guest address space
 * @mm: pointer to the parent mm_struct
 *
 * Returns a guest address space structure.
 */
struct gmap *gmap_alloc(struct mm_struct *mm)
{
	struct gmap *gmap;
	struct page *page;
	unsigned long *table;

	gmap = kzalloc(sizeof(struct gmap), GFP_KERNEL);
	if (!gmap)
		goto out;
	INIT_LIST_HEAD(&gmap->crst_list);
	gmap->mm = mm;
	page = alloc_pages(GFP_KERNEL, ALLOC_ORDER);
	if (!page)
		goto out_free;
	list_add(&page->lru, &gmap->crst_list);
	table = (unsigned long *) page_to_phys(page);
	crst_table_init(table, _REGION1_ENTRY_EMPTY);
	gmap->table = table;
	gmap->asce = _ASCE_TYPE_REGION1 | _ASCE_TABLE_LENGTH |
		     _ASCE_USER_BITS | __pa(table);
	list_add(&gmap->list, &mm->context.gmap_list);
	return gmap;

out_free:
	kfree(gmap);
out:
	return NULL;
}
EXPORT_SYMBOL_GPL(gmap_alloc);

static int gmap_unlink_segment(struct gmap *gmap, unsigned long *table)
{
	struct gmap_pgtable *mp;
	struct gmap_rmap *rmap;
	struct page *page;

	if (*table & _SEGMENT_ENTRY_INV)
		return 0;
	page = pfn_to_page(*table >> PAGE_SHIFT);
	mp = (struct gmap_pgtable *) page->index;
	list_for_each_entry(rmap, &mp->mapper, list) {
		if (rmap->entry != table)
			continue;
		list_del(&rmap->list);
		kfree(rmap);
		break;
	}
	*table = _SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO | mp->vmaddr;
	return 1;
}

static void gmap_flush_tlb(struct gmap *gmap)
{
	if (MACHINE_HAS_IDTE)
		__tlb_flush_idte((unsigned long) gmap->table |
				 _ASCE_TYPE_REGION1);
	else
		__tlb_flush_global();
}

/**
 * gmap_free - free a guest address space
 * @gmap: pointer to the guest address space structure
 */
void gmap_free(struct gmap *gmap)
{
	struct page *page, *next;
	unsigned long *table;
	int i;


	/* Flush tlb. */
	if (MACHINE_HAS_IDTE)
		__tlb_flush_idte((unsigned long) gmap->table |
				 _ASCE_TYPE_REGION1);
	else
		__tlb_flush_global();

	/* Free all segment & region tables. */
	down_read(&gmap->mm->mmap_sem);
	spin_lock(&gmap->mm->page_table_lock);
	list_for_each_entry_safe(page, next, &gmap->crst_list, lru) {
		table = (unsigned long *) page_to_phys(page);
		if ((*table & _REGION_ENTRY_TYPE_MASK) == 0)
			/* Remove gmap rmap structures for segment table. */
			for (i = 0; i < PTRS_PER_PMD; i++, table++)
				gmap_unlink_segment(gmap, table);
		__free_pages(page, ALLOC_ORDER);
	}
	spin_unlock(&gmap->mm->page_table_lock);
	up_read(&gmap->mm->mmap_sem);
	list_del(&gmap->list);
	kfree(gmap);
}
EXPORT_SYMBOL_GPL(gmap_free);

/**
 * gmap_enable - switch primary space to the guest address space
 * @gmap: pointer to the guest address space structure
 */
void gmap_enable(struct gmap *gmap)
{
	S390_lowcore.gmap = (unsigned long) gmap;
}
EXPORT_SYMBOL_GPL(gmap_enable);

/**
 * gmap_disable - switch back to the standard primary address space
 * @gmap: pointer to the guest address space structure
 */
void gmap_disable(struct gmap *gmap)
{
	S390_lowcore.gmap = 0UL;
}
EXPORT_SYMBOL_GPL(gmap_disable);

/*
 * gmap_alloc_table is assumed to be called with mmap_sem held
 */
static int gmap_alloc_table(struct gmap *gmap,
			       unsigned long *table, unsigned long init)
{
	struct page *page;
	unsigned long *new;

	/* since we dont free the gmap table until gmap_free we can unlock */
	spin_unlock(&gmap->mm->page_table_lock);
	page = alloc_pages(GFP_KERNEL, ALLOC_ORDER);
	spin_lock(&gmap->mm->page_table_lock);
	if (!page)
		return -ENOMEM;
	new = (unsigned long *) page_to_phys(page);
	crst_table_init(new, init);
	if (*table & _REGION_ENTRY_INV) {
		list_add(&page->lru, &gmap->crst_list);
		*table = (unsigned long) new | _REGION_ENTRY_LENGTH |
			(*table & _REGION_ENTRY_TYPE_MASK);
	} else
		__free_pages(page, ALLOC_ORDER);
	return 0;
}

/**
 * gmap_unmap_segment - unmap segment from the guest address space
 * @gmap: pointer to the guest address space structure
 * @addr: address in the guest address space
 * @len: length of the memory area to unmap
 *
 * Returns 0 if the unmap succeded, -EINVAL if not.
 */
int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len)
{
	unsigned long *table;
	unsigned long off;
	int flush;

	if ((to | len) & (PMD_SIZE - 1))
		return -EINVAL;
	if (len == 0 || to + len < to)
		return -EINVAL;

	flush = 0;
	down_read(&gmap->mm->mmap_sem);
	spin_lock(&gmap->mm->page_table_lock);
	for (off = 0; off < len; off += PMD_SIZE) {
		/* Walk the guest addr space page table */
		table = gmap->table + (((to + off) >> 53) & 0x7ff);
		if (*table & _REGION_ENTRY_INV)
			goto out;
		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
		table = table + (((to + off) >> 42) & 0x7ff);
		if (*table & _REGION_ENTRY_INV)
			goto out;
		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
		table = table + (((to + off) >> 31) & 0x7ff);
		if (*table & _REGION_ENTRY_INV)
			goto out;
		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
		table = table + (((to + off) >> 20) & 0x7ff);

		/* Clear segment table entry in guest address space. */
		flush |= gmap_unlink_segment(gmap, table);
		*table = _SEGMENT_ENTRY_INV;
	}
out:
	spin_unlock(&gmap->mm->page_table_lock);
	up_read(&gmap->mm->mmap_sem);
	if (flush)
		gmap_flush_tlb(gmap);
	return 0;
}
EXPORT_SYMBOL_GPL(gmap_unmap_segment);

/**
 * gmap_mmap_segment - map a segment to the guest address space
 * @gmap: pointer to the guest address space structure
 * @from: source address in the parent address space
 * @to: target address in the guest address space
 *
 * Returns 0 if the mmap succeded, -EINVAL or -ENOMEM if not.
 */
int gmap_map_segment(struct gmap *gmap, unsigned long from,
		     unsigned long to, unsigned long len)
{
	unsigned long *table;
	unsigned long off;
	int flush;

	if ((from | to | len) & (PMD_SIZE - 1))
		return -EINVAL;
	if (len == 0 || from + len > PGDIR_SIZE ||
	    from + len < from || to + len < to)
		return -EINVAL;

	flush = 0;
	down_read(&gmap->mm->mmap_sem);
	spin_lock(&gmap->mm->page_table_lock);
	for (off = 0; off < len; off += PMD_SIZE) {
		/* Walk the gmap address space page table */
		table = gmap->table + (((to + off) >> 53) & 0x7ff);
		if ((*table & _REGION_ENTRY_INV) &&
		    gmap_alloc_table(gmap, table, _REGION2_ENTRY_EMPTY))
			goto out_unmap;
		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
		table = table + (((to + off) >> 42) & 0x7ff);
		if ((*table & _REGION_ENTRY_INV) &&
		    gmap_alloc_table(gmap, table, _REGION3_ENTRY_EMPTY))
			goto out_unmap;
		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
		table = table + (((to + off) >> 31) & 0x7ff);
		if ((*table & _REGION_ENTRY_INV) &&
		    gmap_alloc_table(gmap, table, _SEGMENT_ENTRY_EMPTY))
			goto out_unmap;
		table = (unsigned long *) (*table & _REGION_ENTRY_ORIGIN);
		table = table + (((to + off) >> 20) & 0x7ff);

		/* Store 'from' address in an invalid segment table entry. */
		flush |= gmap_unlink_segment(gmap, table);
		*table = _SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO | (from + off);
	}
	spin_unlock(&gmap->mm->page_table_lock);
	up_read(&gmap->mm->mmap_sem);
	if (flush)
		gmap_flush_tlb(gmap);
	return 0;

out_unmap:
	spin_unlock(&gmap->mm->page_table_lock);
	up_read(&gmap->mm->mmap_sem);
	gmap_unmap_segment(gmap, to, len);
	return -ENOMEM;
}
EXPORT_SYMBOL_GPL(gmap_map_segment);

/*
 * this function is assumed to be called with mmap_sem held
 */
unsigned long __gmap_fault(unsigned long address, struct gmap *gmap)
{
	unsigned long *table, vmaddr, segment;
	struct mm_struct *mm;
	struct gmap_pgtable *mp;
	struct gmap_rmap *rmap;
	struct vm_area_struct *vma;
	struct page *page;
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;

	current->thread.gmap_addr = address;
	mm = gmap->mm;
	/* Walk the gmap address space page table */
	table = gmap->table + ((address >> 53) & 0x7ff);
	if (unlikely(*table & _REGION_ENTRY_INV))
		return -EFAULT;
	table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
	table = table + ((address >> 42) & 0x7ff);
	if (unlikely(*table & _REGION_ENTRY_INV))
		return -EFAULT;
	table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
	table = table + ((address >> 31) & 0x7ff);
	if (unlikely(*table & _REGION_ENTRY_INV))
		return -EFAULT;
	table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
	table = table + ((address >> 20) & 0x7ff);

	/* Convert the gmap address to an mm address. */
	segment = *table;
	if (likely(!(segment & _SEGMENT_ENTRY_INV))) {
		page = pfn_to_page(segment >> PAGE_SHIFT);
		mp = (struct gmap_pgtable *) page->index;
		return mp->vmaddr | (address & ~PMD_MASK);
	} else if (segment & _SEGMENT_ENTRY_RO) {
		vmaddr = segment & _SEGMENT_ENTRY_ORIGIN;
		vma = find_vma(mm, vmaddr);
		if (!vma || vma->vm_start > vmaddr)
			return -EFAULT;

		/* Walk the parent mm page table */
		pgd = pgd_offset(mm, vmaddr);
		pud = pud_alloc(mm, pgd, vmaddr);
		if (!pud)
			return -ENOMEM;
		pmd = pmd_alloc(mm, pud, vmaddr);
		if (!pmd)
			return -ENOMEM;
		if (!pmd_present(*pmd) &&
		    __pte_alloc(mm, vma, pmd, vmaddr))
			return -ENOMEM;
		/* pmd now points to a valid segment table entry. */
		rmap = kmalloc(sizeof(*rmap), GFP_KERNEL|__GFP_REPEAT);
		if (!rmap)
			return -ENOMEM;
		/* Link gmap segment table entry location to page table. */
		page = pmd_page(*pmd);
		mp = (struct gmap_pgtable *) page->index;
		rmap->entry = table;
		spin_lock(&mm->page_table_lock);
		list_add(&rmap->list, &mp->mapper);
		spin_unlock(&mm->page_table_lock);
		/* Set gmap segment table entry to page table. */
		*table = pmd_val(*pmd) & PAGE_MASK;
		return vmaddr | (address & ~PMD_MASK);
	}
	return -EFAULT;
}

unsigned long gmap_fault(unsigned long address, struct gmap *gmap)
{
	unsigned long rc;

	down_read(&gmap->mm->mmap_sem);
	rc = __gmap_fault(address, gmap);
	up_read(&gmap->mm->mmap_sem);

	return rc;
}
EXPORT_SYMBOL_GPL(gmap_fault);

void gmap_discard(unsigned long from, unsigned long to, struct gmap *gmap)
{

	unsigned long *table, address, size;
	struct vm_area_struct *vma;
	struct gmap_pgtable *mp;
	struct page *page;

	down_read(&gmap->mm->mmap_sem);
	address = from;
	while (address < to) {
		/* Walk the gmap address space page table */
		table = gmap->table + ((address >> 53) & 0x7ff);
		if (unlikely(*table & _REGION_ENTRY_INV)) {
			address = (address + PMD_SIZE) & PMD_MASK;
			continue;
		}
		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
		table = table + ((address >> 42) & 0x7ff);
		if (unlikely(*table & _REGION_ENTRY_INV)) {
			address = (address + PMD_SIZE) & PMD_MASK;
			continue;
		}
		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
		table = table + ((address >> 31) & 0x7ff);
		if (unlikely(*table & _REGION_ENTRY_INV)) {
			address = (address + PMD_SIZE) & PMD_MASK;
			continue;
		}
		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
		table = table + ((address >> 20) & 0x7ff);
		if (unlikely(*table & _SEGMENT_ENTRY_INV)) {
			address = (address + PMD_SIZE) & PMD_MASK;
			continue;
		}
		page = pfn_to_page(*table >> PAGE_SHIFT);
		mp = (struct gmap_pgtable *) page->index;
		vma = find_vma(gmap->mm, mp->vmaddr);
		size = min(to - address, PMD_SIZE - (address & ~PMD_MASK));
		zap_page_range(vma, mp->vmaddr | (address & ~PMD_MASK),
			       size, NULL);
		address = (address + PMD_SIZE) & PMD_MASK;
	}
	up_read(&gmap->mm->mmap_sem);
}
EXPORT_SYMBOL_GPL(gmap_discard);

void gmap_unmap_notifier(struct mm_struct *mm, unsigned long *table)
{
	struct gmap_rmap *rmap, *next;
	struct gmap_pgtable *mp;
	struct page *page;
	int flush;

	flush = 0;
	spin_lock(&mm->page_table_lock);
	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
	mp = (struct gmap_pgtable *) page->index;
	list_for_each_entry_safe(rmap, next, &mp->mapper, list) {
		*rmap->entry =
			_SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO | mp->vmaddr;
		list_del(&rmap->list);
		kfree(rmap);
		flush = 1;
	}
	spin_unlock(&mm->page_table_lock);
	if (flush)
		__tlb_flush_global();
}

static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm,
						    unsigned long vmaddr)
{
	struct page *page;
	unsigned long *table;
	struct gmap_pgtable *mp;

	page = alloc_page(GFP_KERNEL|__GFP_REPEAT);
	if (!page)
		return NULL;
	mp = kmalloc(sizeof(*mp), GFP_KERNEL|__GFP_REPEAT);
	if (!mp) {
		__free_page(page);
		return NULL;
	}
	pgtable_page_ctor(page);
	mp->vmaddr = vmaddr & PMD_MASK;
	INIT_LIST_HEAD(&mp->mapper);
	page->index = (unsigned long) mp;
	atomic_set(&page->_mapcount, 3);
	table = (unsigned long *) page_to_phys(page);
	clear_table(table, _PAGE_TYPE_EMPTY, PAGE_SIZE/2);
	clear_table(table + PTRS_PER_PTE, 0, PAGE_SIZE/2);
	return table;
}

static inline void page_table_free_pgste(unsigned long *table)
{
	struct page *page;
	struct gmap_pgtable *mp;

	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
	mp = (struct gmap_pgtable *) page->index;
	BUG_ON(!list_empty(&mp->mapper));
	pgtable_page_dtor(page);
	atomic_set(&page->_mapcount, -1);
	kfree(mp);
	__free_page(page);
}

#else /* CONFIG_PGSTE */

static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm,
						    unsigned long vmaddr)
{
	return NULL;
}

static inline void page_table_free_pgste(unsigned long *table)
{
}

static inline void gmap_unmap_notifier(struct mm_struct *mm,
					  unsigned long *table)
{
}

#endif /* CONFIG_PGSTE */

static inline unsigned int atomic_xor_bits(atomic_t *v, unsigned int bits)
{
	unsigned int old, new;

	do {
		old = atomic_read(v);
		new = old ^ bits;
	} while (atomic_cmpxchg(v, old, new) != old);
	return new;
}

/*
 * page table entry allocation/free routines.
 */
unsigned long *page_table_alloc(struct mm_struct *mm, unsigned long vmaddr)
{
	unsigned long *uninitialized_var(table);
	struct page *uninitialized_var(page);
	unsigned int mask, bit;

	if (mm_has_pgste(mm))
		return page_table_alloc_pgste(mm, vmaddr);
	/* Allocate fragments of a 4K page as 1K/2K page table */
	spin_lock_bh(&mm->context.list_lock);
	mask = FRAG_MASK;
	if (!list_empty(&mm->context.pgtable_list)) {
		page = list_first_entry(&mm->context.pgtable_list,
					struct page, lru);
		table = (unsigned long *) page_to_phys(page);
		mask = atomic_read(&page->_mapcount);
		mask = mask | (mask >> 4);
	}
	if ((mask & FRAG_MASK) == FRAG_MASK) {
		spin_unlock_bh(&mm->context.list_lock);
		page = alloc_page(GFP_KERNEL|__GFP_REPEAT);
		if (!page)
			return NULL;
		pgtable_page_ctor(page);
		atomic_set(&page->_mapcount, 1);
		table = (unsigned long *) page_to_phys(page);
		clear_table(table, _PAGE_TYPE_EMPTY, PAGE_SIZE);
		spin_lock_bh(&mm->context.list_lock);
		list_add(&page->lru, &mm->context.pgtable_list);
	} else {
		for (bit = 1; mask & bit; bit <<= 1)
			table += PTRS_PER_PTE;
		mask = atomic_xor_bits(&page->_mapcount, bit);
		if ((mask & FRAG_MASK) == FRAG_MASK)
			list_del(&page->lru);
	}
	spin_unlock_bh(&mm->context.list_lock);
	return table;
}

void page_table_free(struct mm_struct *mm, unsigned long *table)
{
	struct page *page;
	unsigned int bit, mask;

	if (mm_has_pgste(mm)) {
		gmap_unmap_notifier(mm, table);
		return page_table_free_pgste(table);
	}
	/* Free 1K/2K page table fragment of a 4K page */
	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
	bit = 1 << ((__pa(table) & ~PAGE_MASK)/(PTRS_PER_PTE*sizeof(pte_t)));
	spin_lock_bh(&mm->context.list_lock);
	if ((atomic_read(&page->_mapcount) & FRAG_MASK) != FRAG_MASK)
		list_del(&page->lru);
	mask = atomic_xor_bits(&page->_mapcount, bit);
	if (mask & FRAG_MASK)
		list_add(&page->lru, &mm->context.pgtable_list);
	spin_unlock_bh(&mm->context.list_lock);
	if (mask == 0) {
		pgtable_page_dtor(page);
		atomic_set(&page->_mapcount, -1);
		__free_page(page);
	}
}

static void __page_table_free_rcu(void *table, unsigned bit)
{
	struct page *page;

	if (bit == FRAG_MASK)
		return page_table_free_pgste(table);
	/* Free 1K/2K page table fragment of a 4K page */
	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
	if (atomic_xor_bits(&page->_mapcount, bit) == 0) {
		pgtable_page_dtor(page);
		atomic_set(&page->_mapcount, -1);
		__free_page(page);
	}
}

void page_table_free_rcu(struct mmu_gather *tlb, unsigned long *table)
{
	struct mm_struct *mm;
	struct page *page;
	unsigned int bit, mask;

	mm = tlb->mm;
	if (mm_has_pgste(mm)) {
		gmap_unmap_notifier(mm, table);
		table = (unsigned long *) (__pa(table) | FRAG_MASK);
		tlb_remove_table(tlb, table);
		return;
	}
	bit = 1 << ((__pa(table) & ~PAGE_MASK) / (PTRS_PER_PTE*sizeof(pte_t)));
	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
	spin_lock_bh(&mm->context.list_lock);
	if ((atomic_read(&page->_mapcount) & FRAG_MASK) != FRAG_MASK)
		list_del(&page->lru);
	mask = atomic_xor_bits(&page->_mapcount, bit | (bit << 4));
	if (mask & FRAG_MASK)
		list_add_tail(&page->lru, &mm->context.pgtable_list);
	spin_unlock_bh(&mm->context.list_lock);
	table = (unsigned long *) (__pa(table) | (bit << 4));
	tlb_remove_table(tlb, table);
}

void __tlb_remove_table(void *_table)
{
	const unsigned long mask = (FRAG_MASK << 4) | FRAG_MASK;
	void *table = (void *)((unsigned long) _table & ~mask);
	unsigned type = (unsigned long) _table & mask;

	if (type)
		__page_table_free_rcu(table, type);
	else
		free_pages((unsigned long) table, ALLOC_ORDER);
}

static void tlb_remove_table_smp_sync(void *arg)
{
	/* Simply deliver the interrupt */
}

static void tlb_remove_table_one(void *table)
{
	/*
	 * This isn't an RCU grace period and hence the page-tables cannot be
	 * assumed to be actually RCU-freed.
	 *
	 * It is however sufficient for software page-table walkers that rely
	 * on IRQ disabling. See the comment near struct mmu_table_batch.
	 */
	smp_call_function(tlb_remove_table_smp_sync, NULL, 1);
	__tlb_remove_table(table);
}

static void tlb_remove_table_rcu(struct rcu_head *head)
{
	struct mmu_table_batch *batch;
	int i;

	batch = container_of(head, struct mmu_table_batch, rcu);

	for (i = 0; i < batch->nr; i++)
		__tlb_remove_table(batch->tables[i]);

	free_page((unsigned long)batch);
}

void tlb_table_flush(struct mmu_gather *tlb)
{
	struct mmu_table_batch **batch = &tlb->batch;

	if (*batch) {
		__tlb_flush_mm(tlb->mm);
		call_rcu_sched(&(*batch)->rcu, tlb_remove_table_rcu);
		*batch = NULL;
	}
}

void tlb_remove_table(struct mmu_gather *tlb, void *table)
{
	struct mmu_table_batch **batch = &tlb->batch;

	if (*batch == NULL) {
		*batch = (struct mmu_table_batch *)
			__get_free_page(GFP_NOWAIT | __GFP_NOWARN);
		if (*batch == NULL) {
			__tlb_flush_mm(tlb->mm);
			tlb_remove_table_one(table);
			return;
		}
		(*batch)->nr = 0;
	}
	(*batch)->tables[(*batch)->nr++] = table;
	if ((*batch)->nr == MAX_TABLE_BATCH)
		tlb_table_flush(tlb);
}

#ifdef CONFIG_TRANSPARENT_HUGEPAGE
void thp_split_vma(struct vm_area_struct *vma)
{
	unsigned long addr;
	struct page *page;

	for (addr = vma->vm_start; addr < vma->vm_end; addr += PAGE_SIZE) {
		page = follow_page(vma, addr, FOLL_SPLIT);
	}
}

void thp_split_mm(struct mm_struct *mm)
{
	struct vm_area_struct *vma = mm->mmap;

	while (vma != NULL) {
		thp_split_vma(vma);
		vma->vm_flags &= ~VM_HUGEPAGE;
		vma->vm_flags |= VM_NOHUGEPAGE;
		vma = vma->vm_next;
	}
}
#endif /* CONFIG_TRANSPARENT_HUGEPAGE */

/*
 * switch on pgstes for its userspace process (for kvm)
 */
int s390_enable_sie(void)
{
	struct task_struct *tsk = current;
	struct mm_struct *mm, *old_mm;

	/* Do we have switched amode? If no, we cannot do sie */
	if (s390_user_mode == HOME_SPACE_MODE)
		return -EINVAL;

	/* Do we have pgstes? if yes, we are done */
	if (mm_has_pgste(tsk->mm))
		return 0;

	/* lets check if we are allowed to replace the mm */
	task_lock(tsk);
	if (!tsk->mm || atomic_read(&tsk->mm->mm_users) > 1 ||
#ifdef CONFIG_AIO
	    !hlist_empty(&tsk->mm->ioctx_list) ||
#endif
	    tsk->mm != tsk->active_mm) {
		task_unlock(tsk);
		return -EINVAL;
	}
	task_unlock(tsk);

	/* we copy the mm and let dup_mm create the page tables with_pgstes */
	tsk->mm->context.alloc_pgste = 1;
	/* make sure that both mms have a correct rss state */
	sync_mm_rss(tsk->mm);
	mm = dup_mm(tsk);
	tsk->mm->context.alloc_pgste = 0;
	if (!mm)
		return -ENOMEM;

#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	/* split thp mappings and disable thp for future mappings */
	thp_split_mm(mm);
	mm->def_flags |= VM_NOHUGEPAGE;
#endif

	/* Now lets check again if something happened */
	task_lock(tsk);
	if (!tsk->mm || atomic_read(&tsk->mm->mm_users) > 1 ||
#ifdef CONFIG_AIO
	    !hlist_empty(&tsk->mm->ioctx_list) ||
#endif
	    tsk->mm != tsk->active_mm) {
		mmput(mm);
		task_unlock(tsk);
		return -EINVAL;
	}

	/* ok, we are alone. No ptrace, no threads, etc. */
	old_mm = tsk->mm;
	tsk->mm = tsk->active_mm = mm;
	preempt_disable();
	update_mm(mm, tsk);
	atomic_inc(&mm->context.attach_count);
	atomic_dec(&old_mm->context.attach_count);
	cpumask_set_cpu(smp_processor_id(), mm_cpumask(mm));
	preempt_enable();
	task_unlock(tsk);
	mmput(old_mm);
	return 0;
}
EXPORT_SYMBOL_GPL(s390_enable_sie);

#if defined(CONFIG_DEBUG_PAGEALLOC) && defined(CONFIG_HIBERNATION)
bool kernel_page_present(struct page *page)
{
	unsigned long addr;
	int cc;

	addr = page_to_phys(page);
	asm volatile(
		"	lra	%1,0(%1)\n"
		"	ipm	%0\n"
		"	srl	%0,28"
		: "=d" (cc), "+a" (addr) : : "cc");
	return cc == 0;
}
#endif /* CONFIG_HIBERNATION && CONFIG_DEBUG_PAGEALLOC */

#ifdef CONFIG_TRANSPARENT_HUGEPAGE
int pmdp_clear_flush_young(struct vm_area_struct *vma, unsigned long address,
			   pmd_t *pmdp)
{
	VM_BUG_ON(address & ~HPAGE_PMD_MASK);
	/* No need to flush TLB
	 * On s390 reference bits are in storage key and never in TLB */
	return pmdp_test_and_clear_young(vma, address, pmdp);
}

int pmdp_set_access_flags(struct vm_area_struct *vma,
			  unsigned long address, pmd_t *pmdp,
			  pmd_t entry, int dirty)
{
	VM_BUG_ON(address & ~HPAGE_PMD_MASK);

	if (pmd_same(*pmdp, entry))
		return 0;
	pmdp_invalidate(vma, address, pmdp);
	set_pmd_at(vma->vm_mm, address, pmdp, entry);
	return 1;
}

static void pmdp_splitting_flush_sync(void *arg)
{
	/* Simply deliver the interrupt */
}

void pmdp_splitting_flush(struct vm_area_struct *vma, unsigned long address,
			  pmd_t *pmdp)
{
	VM_BUG_ON(address & ~HPAGE_PMD_MASK);
	if (!test_and_set_bit(_SEGMENT_ENTRY_SPLIT_BIT,
			      (unsigned long *) pmdp)) {
		/* need to serialize against gup-fast (IRQ disabled) */
		smp_call_function(pmdp_splitting_flush_sync, NULL, 1);
	}
}

void pgtable_trans_huge_deposit(struct mm_struct *mm, pgtable_t pgtable)
{
	struct list_head *lh = (struct list_head *) pgtable;

	assert_spin_locked(&mm->page_table_lock);

	/* FIFO */
	if (!mm->pmd_huge_pte)
		INIT_LIST_HEAD(lh);
	else
		list_add(lh, (struct list_head *) mm->pmd_huge_pte);
	mm->pmd_huge_pte = pgtable;
}

pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm)
{
	struct list_head *lh;
	pgtable_t pgtable;
	pte_t *ptep;

	assert_spin_locked(&mm->page_table_lock);

	/* FIFO */
	pgtable = mm->pmd_huge_pte;
	lh = (struct list_head *) pgtable;
	if (list_empty(lh))
		mm->pmd_huge_pte = NULL;
	else {
		mm->pmd_huge_pte = (pgtable_t) lh->next;
		list_del(lh);
	}
	ptep = (pte_t *) pgtable;
	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
	ptep++;
	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
	return pgtable;
}
#endif /* CONFIG_TRANSPARENT_HUGEPAGE */