aboutsummaryrefslogtreecommitdiff
path: root/arch/tile/mm/hugetlbpage.c
blob: e514899e1100319dc83fe69530f1aad67b17ceea (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
/*
 * Copyright 2010 Tilera Corporation. All Rights Reserved.
 *
 *   This program is free software; you can redistribute it and/or
 *   modify it under the terms of the GNU General Public License
 *   as published by the Free Software Foundation, version 2.
 *
 *   This program is distributed in the hope that it will be useful, but
 *   WITHOUT ANY WARRANTY; without even the implied warranty of
 *   MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
 *   NON INFRINGEMENT.  See the GNU General Public License for
 *   more details.
 *
 * TILE Huge TLB Page Support for Kernel.
 * Taken from i386 hugetlb implementation:
 * Copyright (C) 2002, Rohit Seth <rohit.seth@intel.com>
 */

#include <linux/init.h>
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/hugetlb.h>
#include <linux/pagemap.h>
#include <linux/slab.h>
#include <linux/err.h>
#include <linux/sysctl.h>
#include <linux/mman.h>
#include <asm/tlb.h>
#include <asm/tlbflush.h>
#include <asm/setup.h>

#ifdef CONFIG_HUGETLB_SUPER_PAGES

/*
 * Provide an additional huge page size (in addition to the regular default
 * huge page size) if no "hugepagesz" arguments are specified.
 * Note that it must be smaller than the default huge page size so
 * that it's possible to allocate them on demand from the buddy allocator.
 * You can change this to 64K (on a 16K build), 256K, 1M, or 4M,
 * or not define it at all.
 */
#define ADDITIONAL_HUGE_SIZE (1024 * 1024UL)

/* "Extra" page-size multipliers, one per level of the page table. */
int huge_shift[HUGE_SHIFT_ENTRIES] = {
#ifdef ADDITIONAL_HUGE_SIZE
#define ADDITIONAL_HUGE_SHIFT __builtin_ctzl(ADDITIONAL_HUGE_SIZE / PAGE_SIZE)
	[HUGE_SHIFT_PAGE] = ADDITIONAL_HUGE_SHIFT
#endif
};

#endif

pte_t *huge_pte_alloc(struct mm_struct *mm,
		      unsigned long addr, unsigned long sz)
{
	pgd_t *pgd;
	pud_t *pud;

	addr &= -sz;   /* Mask off any low bits in the address. */

	pgd = pgd_offset(mm, addr);
	pud = pud_alloc(mm, pgd, addr);

#ifdef CONFIG_HUGETLB_SUPER_PAGES
	if (sz >= PGDIR_SIZE) {
		BUG_ON(sz != PGDIR_SIZE &&
		       sz != PGDIR_SIZE << huge_shift[HUGE_SHIFT_PGDIR]);
		return (pte_t *)pud;
	} else {
		pmd_t *pmd = pmd_alloc(mm, pud, addr);
		if (sz >= PMD_SIZE) {
			BUG_ON(sz != PMD_SIZE &&
			       sz != (PMD_SIZE << huge_shift[HUGE_SHIFT_PMD]));
			return (pte_t *)pmd;
		}
		else {
			if (sz != PAGE_SIZE << huge_shift[HUGE_SHIFT_PAGE])
				panic("Unexpected page size %#lx\n", sz);
			return pte_alloc_map(mm, NULL, pmd, addr);
		}
	}
#else
	BUG_ON(sz != PMD_SIZE);
	return (pte_t *) pmd_alloc(mm, pud, addr);
#endif
}

static pte_t *get_pte(pte_t *base, int index, int level)
{
	pte_t *ptep = base + index;
#ifdef CONFIG_HUGETLB_SUPER_PAGES
	if (!pte_present(*ptep) && huge_shift[level] != 0) {
		unsigned long mask = -1UL << huge_shift[level];
		pte_t *super_ptep = base + (index & mask);
		pte_t pte = *super_ptep;
		if (pte_present(pte) && pte_super(pte))
			ptep = super_ptep;
	}
#endif
	return ptep;
}

pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
{
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
#ifdef CONFIG_HUGETLB_SUPER_PAGES
	pte_t *pte;
#endif

	/* Get the top-level page table entry. */
	pgd = (pgd_t *)get_pte((pte_t *)mm->pgd, pgd_index(addr), 0);

	/* We don't have four levels. */
	pud = pud_offset(pgd, addr);
#ifndef __PAGETABLE_PUD_FOLDED
# error support fourth page table level
#endif
	if (!pud_present(*pud))
		return NULL;

	/* Check for an L0 huge PTE, if we have three levels. */
#ifndef __PAGETABLE_PMD_FOLDED
	if (pud_huge(*pud))
		return (pte_t *)pud;

	pmd = (pmd_t *)get_pte((pte_t *)pud_page_vaddr(*pud),
			       pmd_index(addr), 1);
	if (!pmd_present(*pmd))
		return NULL;
#else
	pmd = pmd_offset(pud, addr);
#endif

	/* Check for an L1 huge PTE. */
	if (pmd_huge(*pmd))
		return (pte_t *)pmd;

#ifdef CONFIG_HUGETLB_SUPER_PAGES
	/* Check for an L2 huge PTE. */
	pte = get_pte((pte_t *)pmd_page_vaddr(*pmd), pte_index(addr), 2);
	if (!pte_present(*pte))
		return NULL;
	if (pte_super(*pte))
		return pte;
#endif

	return NULL;
}

struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address,
			      int write)
{
	return ERR_PTR(-EINVAL);
}

int pmd_huge(pmd_t pmd)
{
	return !!(pmd_val(pmd) & _PAGE_HUGE_PAGE);
}

int pud_huge(pud_t pud)
{
	return !!(pud_val(pud) & _PAGE_HUGE_PAGE);
}

struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address,
			     pmd_t *pmd, int write)
{
	struct page *page;

	page = pte_page(*(pte_t *)pmd);
	if (page)
		page += ((address & ~PMD_MASK) >> PAGE_SHIFT);
	return page;
}

struct page *follow_huge_pud(struct mm_struct *mm, unsigned long address,
			     pud_t *pud, int write)
{
	struct page *page;

	page = pte_page(*(pte_t *)pud);
	if (page)
		page += ((address & ~PUD_MASK) >> PAGE_SHIFT);
	return page;
}

int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep)
{
	return 0;
}

#ifdef HAVE_ARCH_HUGETLB_UNMAPPED_AREA
static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file,
		unsigned long addr, unsigned long len,
		unsigned long pgoff, unsigned long flags)
{
	struct hstate *h = hstate_file(file);
	struct vm_unmapped_area_info info;

	info.flags = 0;
	info.length = len;
	info.low_limit = TASK_UNMAPPED_BASE;
	info.high_limit = TASK_SIZE;
	info.align_mask = PAGE_MASK & ~huge_page_mask(h);
	info.align_offset = 0;
	return vm_unmapped_area(&info);
}

static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file,
		unsigned long addr0, unsigned long len,
		unsigned long pgoff, unsigned long flags)
{
	struct hstate *h = hstate_file(file);
	struct vm_unmapped_area_info info;
	unsigned long addr;

	info.flags = VM_UNMAPPED_AREA_TOPDOWN;
	info.length = len;
	info.low_limit = PAGE_SIZE;
	info.high_limit = current->mm->mmap_base;
	info.align_mask = PAGE_MASK & ~huge_page_mask(h);
	info.align_offset = 0;
	addr = vm_unmapped_area(&info);

	/*
	 * A failed mmap() very likely causes application failure,
	 * so fall back to the bottom-up function here. This scenario
	 * can happen with large stack limits and large mmap()
	 * allocations.
	 */
	if (addr & ~PAGE_MASK) {
		VM_BUG_ON(addr != -ENOMEM);
		info.flags = 0;
		info.low_limit = TASK_UNMAPPED_BASE;
		info.high_limit = TASK_SIZE;
		addr = vm_unmapped_area(&info);
	}

	return addr;
}

unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
		unsigned long len, unsigned long pgoff, unsigned long flags)
{
	struct hstate *h = hstate_file(file);
	struct mm_struct *mm = current->mm;
	struct vm_area_struct *vma;

	if (len & ~huge_page_mask(h))
		return -EINVAL;
	if (len > TASK_SIZE)
		return -ENOMEM;

	if (flags & MAP_FIXED) {
		if (prepare_hugepage_range(file, addr, len))
			return -EINVAL;
		return addr;
	}

	if (addr) {
		addr = ALIGN(addr, huge_page_size(h));
		vma = find_vma(mm, addr);
		if (TASK_SIZE - len >= addr &&
		    (!vma || addr + len <= vma->vm_start))
			return addr;
	}
	if (current->mm->get_unmapped_area == arch_get_unmapped_area)
		return hugetlb_get_unmapped_area_bottomup(file, addr, len,
				pgoff, flags);
	else
		return hugetlb_get_unmapped_area_topdown(file, addr, len,
				pgoff, flags);
}
#endif /* HAVE_ARCH_HUGETLB_UNMAPPED_AREA */

#ifdef CONFIG_HUGETLB_SUPER_PAGES
static __init int __setup_hugepagesz(unsigned long ps)
{
	int log_ps = __builtin_ctzl(ps);
	int level, base_shift;

	if ((1UL << log_ps) != ps || (log_ps & 1) != 0) {
		pr_warn("Not enabling %ld byte huge pages;"
			" must be a power of four.\n", ps);
		return -EINVAL;
	}

	if (ps > 64*1024*1024*1024UL) {
		pr_warn("Not enabling %ld MB huge pages;"
			" largest legal value is 64 GB .\n", ps >> 20);
		return -EINVAL;
	} else if (ps >= PUD_SIZE) {
		static long hv_jpage_size;
		if (hv_jpage_size == 0)
			hv_jpage_size = hv_sysconf(HV_SYSCONF_PAGE_SIZE_JUMBO);
		if (hv_jpage_size != PUD_SIZE) {
			pr_warn("Not enabling >= %ld MB huge pages:"
				" hypervisor reports size %ld\n",
				PUD_SIZE >> 20, hv_jpage_size);
			return -EINVAL;
		}
		level = 0;
		base_shift = PUD_SHIFT;
	} else if (ps >= PMD_SIZE) {
		level = 1;
		base_shift = PMD_SHIFT;
	} else if (ps > PAGE_SIZE) {
		level = 2;
		base_shift = PAGE_SHIFT;
	} else {
		pr_err("hugepagesz: huge page size %ld too small\n", ps);
		return -EINVAL;
	}

	if (log_ps != base_shift) {
		int shift_val = log_ps - base_shift;
		if (huge_shift[level] != 0) {
			int old_shift = base_shift + huge_shift[level];
			pr_warn("Not enabling %ld MB huge pages;"
				" already have size %ld MB.\n",
				ps >> 20, (1UL << old_shift) >> 20);
			return -EINVAL;
		}
		if (hv_set_pte_super_shift(level, shift_val) != 0) {
			pr_warn("Not enabling %ld MB huge pages;"
				" no hypervisor support.\n", ps >> 20);
			return -EINVAL;
		}
		printk(KERN_DEBUG "Enabled %ld MB huge pages\n", ps >> 20);
		huge_shift[level] = shift_val;
	}

	hugetlb_add_hstate(log_ps - PAGE_SHIFT);

	return 0;
}

static bool saw_hugepagesz;

static __init int setup_hugepagesz(char *opt)
{
	if (!saw_hugepagesz) {
		saw_hugepagesz = true;
		memset(huge_shift, 0, sizeof(huge_shift));
	}
	return __setup_hugepagesz(memparse(opt, NULL));
}
__setup("hugepagesz=", setup_hugepagesz);

#ifdef ADDITIONAL_HUGE_SIZE
/*
 * Provide an additional huge page size if no "hugepagesz" args are given.
 * In that case, all the cores have properly set up their hv super_shift
 * already, but we need to notify the hugetlb code to enable the
 * new huge page size from the Linux point of view.
 */
static __init int add_default_hugepagesz(void)
{
	if (!saw_hugepagesz) {
		BUILD_BUG_ON(ADDITIONAL_HUGE_SIZE >= PMD_SIZE ||
			     ADDITIONAL_HUGE_SIZE <= PAGE_SIZE);
		BUILD_BUG_ON((PAGE_SIZE << ADDITIONAL_HUGE_SHIFT) !=
			     ADDITIONAL_HUGE_SIZE);
		BUILD_BUG_ON(ADDITIONAL_HUGE_SHIFT & 1);
		hugetlb_add_hstate(ADDITIONAL_HUGE_SHIFT);
	}
	return 0;
}
arch_initcall(add_default_hugepagesz);
#endif

#endif /* CONFIG_HUGETLB_SUPER_PAGES */