Commit | Line | Data |
---|---|---|
c64be2bb MS |
1 | /* |
2 | * Contiguous Memory Allocator for DMA mapping framework | |
3 | * Copyright (c) 2010-2011 by Samsung Electronics. | |
4 | * Written by: | |
5 | * Marek Szyprowski <m.szyprowski@samsung.com> | |
6 | * Michal Nazarewicz <mina86@mina86.com> | |
7 | * | |
8 | * This program is free software; you can redistribute it and/or | |
9 | * modify it under the terms of the GNU General Public License as | |
10 | * published by the Free Software Foundation; either version 2 of the | |
11 | * License or (at your optional) any later version of the license. | |
12 | */ | |
13 | ||
14 | #define pr_fmt(fmt) "cma: " fmt | |
15 | ||
16 | #ifdef CONFIG_CMA_DEBUG | |
17 | #ifndef DEBUG | |
18 | # define DEBUG | |
19 | #endif | |
20 | #endif | |
21 | ||
22 | #include <asm/page.h> | |
23 | #include <asm/dma-contiguous.h> | |
24 | ||
25 | #include <linux/memblock.h> | |
26 | #include <linux/err.h> | |
27 | #include <linux/mm.h> | |
28 | #include <linux/mutex.h> | |
29 | #include <linux/page-isolation.h> | |
446c82fc | 30 | #include <linux/sizes.h> |
c64be2bb MS |
31 | #include <linux/slab.h> |
32 | #include <linux/swap.h> | |
33 | #include <linux/mm_types.h> | |
34 | #include <linux/dma-contiguous.h> | |
35 | ||
c64be2bb MS |
36 | struct cma { |
37 | unsigned long base_pfn; | |
38 | unsigned long count; | |
39 | unsigned long *bitmap; | |
40 | }; | |
41 | ||
42 | struct cma *dma_contiguous_default_area; | |
43 | ||
44 | #ifdef CONFIG_CMA_SIZE_MBYTES | |
45 | #define CMA_SIZE_MBYTES CONFIG_CMA_SIZE_MBYTES | |
46 | #else | |
47 | #define CMA_SIZE_MBYTES 0 | |
48 | #endif | |
49 | ||
50 | /* | |
51 | * Default global CMA area size can be defined in kernel's .config. | |
52 | * This is usefull mainly for distro maintainers to create a kernel | |
53 | * that works correctly for most supported systems. | |
54 | * The size can be set in bytes or as a percentage of the total memory | |
55 | * in the system. | |
56 | * | |
57 | * Users, who want to set the size of global CMA area for their system | |
58 | * should use cma= kernel parameter. | |
59 | */ | |
4009793e VA |
60 | static const phys_addr_t size_bytes = CMA_SIZE_MBYTES * SZ_1M; |
61 | static phys_addr_t size_cmdline = -1; | |
c64be2bb MS |
62 | |
63 | static int __init early_cma(char *p) | |
64 | { | |
65 | pr_debug("%s(%s)\n", __func__, p); | |
66 | size_cmdline = memparse(p, &p); | |
67 | return 0; | |
68 | } | |
69 | early_param("cma", early_cma); | |
70 | ||
71 | #ifdef CONFIG_CMA_SIZE_PERCENTAGE | |
72 | ||
4009793e | 73 | static phys_addr_t __init __maybe_unused cma_early_percent_memory(void) |
c64be2bb MS |
74 | { |
75 | struct memblock_region *reg; | |
76 | unsigned long total_pages = 0; | |
77 | ||
78 | /* | |
79 | * We cannot use memblock_phys_mem_size() here, because | |
80 | * memblock_analyze() has not been called yet. | |
81 | */ | |
82 | for_each_memblock(memory, reg) | |
83 | total_pages += memblock_region_memory_end_pfn(reg) - | |
84 | memblock_region_memory_base_pfn(reg); | |
85 | ||
86 | return (total_pages * CONFIG_CMA_SIZE_PERCENTAGE / 100) << PAGE_SHIFT; | |
87 | } | |
88 | ||
89 | #else | |
90 | ||
4009793e | 91 | static inline __maybe_unused phys_addr_t cma_early_percent_memory(void) |
c64be2bb MS |
92 | { |
93 | return 0; | |
94 | } | |
95 | ||
96 | #endif | |
97 | ||
98 | /** | |
99 | * dma_contiguous_reserve() - reserve area for contiguous memory handling | |
100 | * @limit: End address of the reserved memory (optional, 0 for any). | |
101 | * | |
102 | * This function reserves memory from early allocator. It should be | |
103 | * called by arch specific code once the early allocator (memblock or bootmem) | |
104 | * has been activated and all other subsystems have already allocated/reserved | |
105 | * memory. | |
106 | */ | |
107 | void __init dma_contiguous_reserve(phys_addr_t limit) | |
108 | { | |
4009793e | 109 | phys_addr_t selected_size = 0; |
c64be2bb MS |
110 | |
111 | pr_debug("%s(limit %08lx)\n", __func__, (unsigned long)limit); | |
112 | ||
113 | if (size_cmdline != -1) { | |
114 | selected_size = size_cmdline; | |
115 | } else { | |
116 | #ifdef CONFIG_CMA_SIZE_SEL_MBYTES | |
117 | selected_size = size_bytes; | |
118 | #elif defined(CONFIG_CMA_SIZE_SEL_PERCENTAGE) | |
119 | selected_size = cma_early_percent_memory(); | |
120 | #elif defined(CONFIG_CMA_SIZE_SEL_MIN) | |
121 | selected_size = min(size_bytes, cma_early_percent_memory()); | |
122 | #elif defined(CONFIG_CMA_SIZE_SEL_MAX) | |
123 | selected_size = max(size_bytes, cma_early_percent_memory()); | |
124 | #endif | |
125 | } | |
126 | ||
127 | if (selected_size) { | |
128 | pr_debug("%s: reserving %ld MiB for global area\n", __func__, | |
4009793e | 129 | (unsigned long)selected_size / SZ_1M); |
c64be2bb MS |
130 | |
131 | dma_declare_contiguous(NULL, selected_size, 0, limit); | |
132 | } | |
133 | }; | |
134 | ||
135 | static DEFINE_MUTEX(cma_mutex); | |
136 | ||
137 | static __init int cma_activate_area(unsigned long base_pfn, unsigned long count) | |
138 | { | |
139 | unsigned long pfn = base_pfn; | |
140 | unsigned i = count >> pageblock_order; | |
141 | struct zone *zone; | |
142 | ||
143 | WARN_ON_ONCE(!pfn_valid(pfn)); | |
144 | zone = page_zone(pfn_to_page(pfn)); | |
145 | ||
146 | do { | |
147 | unsigned j; | |
148 | base_pfn = pfn; | |
149 | for (j = pageblock_nr_pages; j; --j, pfn++) { | |
150 | WARN_ON_ONCE(!pfn_valid(pfn)); | |
151 | if (page_zone(pfn_to_page(pfn)) != zone) | |
152 | return -EINVAL; | |
153 | } | |
154 | init_cma_reserved_pageblock(pfn_to_page(base_pfn)); | |
155 | } while (--i); | |
156 | return 0; | |
157 | } | |
158 | ||
159 | static __init struct cma *cma_create_area(unsigned long base_pfn, | |
160 | unsigned long count) | |
161 | { | |
162 | int bitmap_size = BITS_TO_LONGS(count) * sizeof(long); | |
163 | struct cma *cma; | |
164 | int ret = -ENOMEM; | |
165 | ||
166 | pr_debug("%s(base %08lx, count %lx)\n", __func__, base_pfn, count); | |
167 | ||
168 | cma = kmalloc(sizeof *cma, GFP_KERNEL); | |
169 | if (!cma) | |
170 | return ERR_PTR(-ENOMEM); | |
171 | ||
172 | cma->base_pfn = base_pfn; | |
173 | cma->count = count; | |
174 | cma->bitmap = kzalloc(bitmap_size, GFP_KERNEL); | |
175 | ||
176 | if (!cma->bitmap) | |
177 | goto no_mem; | |
178 | ||
179 | ret = cma_activate_area(base_pfn, count); | |
180 | if (ret) | |
181 | goto error; | |
182 | ||
183 | pr_debug("%s: returned %p\n", __func__, (void *)cma); | |
184 | return cma; | |
185 | ||
186 | error: | |
187 | kfree(cma->bitmap); | |
188 | no_mem: | |
189 | kfree(cma); | |
190 | return ERR_PTR(ret); | |
191 | } | |
192 | ||
193 | static struct cma_reserved { | |
194 | phys_addr_t start; | |
195 | unsigned long size; | |
196 | struct device *dev; | |
197 | } cma_reserved[MAX_CMA_AREAS] __initdata; | |
198 | static unsigned cma_reserved_count __initdata; | |
199 | ||
200 | static int __init cma_init_reserved_areas(void) | |
201 | { | |
202 | struct cma_reserved *r = cma_reserved; | |
203 | unsigned i = cma_reserved_count; | |
204 | ||
205 | pr_debug("%s()\n", __func__); | |
206 | ||
207 | for (; i; --i, ++r) { | |
208 | struct cma *cma; | |
209 | cma = cma_create_area(PFN_DOWN(r->start), | |
210 | r->size >> PAGE_SHIFT); | |
211 | if (!IS_ERR(cma)) | |
212 | dev_set_cma_area(r->dev, cma); | |
213 | } | |
214 | return 0; | |
215 | } | |
216 | core_initcall(cma_init_reserved_areas); | |
217 | ||
218 | /** | |
219 | * dma_declare_contiguous() - reserve area for contiguous memory handling | |
220 | * for particular device | |
221 | * @dev: Pointer to device structure. | |
222 | * @size: Size of the reserved memory. | |
223 | * @base: Start address of the reserved memory (optional, 0 for any). | |
224 | * @limit: End address of the reserved memory (optional, 0 for any). | |
225 | * | |
226 | * This function reserves memory for specified device. It should be | |
227 | * called by board specific code when early allocator (memblock or bootmem) | |
228 | * is still activate. | |
229 | */ | |
4009793e | 230 | int __init dma_declare_contiguous(struct device *dev, phys_addr_t size, |
c64be2bb MS |
231 | phys_addr_t base, phys_addr_t limit) |
232 | { | |
233 | struct cma_reserved *r = &cma_reserved[cma_reserved_count]; | |
4009793e | 234 | phys_addr_t alignment; |
c64be2bb MS |
235 | |
236 | pr_debug("%s(size %lx, base %08lx, limit %08lx)\n", __func__, | |
237 | (unsigned long)size, (unsigned long)base, | |
238 | (unsigned long)limit); | |
239 | ||
240 | /* Sanity checks */ | |
241 | if (cma_reserved_count == ARRAY_SIZE(cma_reserved)) { | |
242 | pr_err("Not enough slots for CMA reserved regions!\n"); | |
243 | return -ENOSPC; | |
244 | } | |
245 | ||
246 | if (!size) | |
247 | return -EINVAL; | |
248 | ||
249 | /* Sanitise input arguments */ | |
7ce9bf1f | 250 | alignment = PAGE_SIZE << max(MAX_ORDER - 1, pageblock_order); |
c64be2bb MS |
251 | base = ALIGN(base, alignment); |
252 | size = ALIGN(size, alignment); | |
253 | limit &= ~(alignment - 1); | |
254 | ||
255 | /* Reserve memory */ | |
256 | if (base) { | |
257 | if (memblock_is_region_reserved(base, size) || | |
258 | memblock_reserve(base, size) < 0) { | |
259 | base = -EBUSY; | |
260 | goto err; | |
261 | } | |
262 | } else { | |
263 | /* | |
264 | * Use __memblock_alloc_base() since | |
265 | * memblock_alloc_base() panic()s. | |
266 | */ | |
267 | phys_addr_t addr = __memblock_alloc_base(size, alignment, limit); | |
268 | if (!addr) { | |
269 | base = -ENOMEM; | |
270 | goto err; | |
c64be2bb MS |
271 | } else { |
272 | base = addr; | |
273 | } | |
274 | } | |
275 | ||
276 | /* | |
277 | * Each reserved area must be initialised later, when more kernel | |
278 | * subsystems (like slab allocator) are available. | |
279 | */ | |
280 | r->start = base; | |
281 | r->size = size; | |
282 | r->dev = dev; | |
283 | cma_reserved_count++; | |
4009793e | 284 | pr_info("CMA: reserved %ld MiB at %08lx\n", (unsigned long)size / SZ_1M, |
c64be2bb MS |
285 | (unsigned long)base); |
286 | ||
287 | /* Architecture specific contiguous memory fixup. */ | |
288 | dma_contiguous_early_fixup(base, size); | |
289 | return 0; | |
290 | err: | |
4009793e | 291 | pr_err("CMA: failed to reserve %ld MiB\n", (unsigned long)size / SZ_1M); |
c64be2bb MS |
292 | return base; |
293 | } | |
294 | ||
295 | /** | |
296 | * dma_alloc_from_contiguous() - allocate pages from contiguous area | |
297 | * @dev: Pointer to device for which the allocation is performed. | |
298 | * @count: Requested number of pages. | |
299 | * @align: Requested alignment of pages (in PAGE_SIZE order). | |
300 | * | |
301 | * This function allocates memory buffer for specified device. It uses | |
302 | * device specific contiguous memory area if available or the default | |
303 | * global one. Requires architecture specific get_dev_cma_area() helper | |
304 | * function. | |
305 | */ | |
306 | struct page *dma_alloc_from_contiguous(struct device *dev, int count, | |
307 | unsigned int align) | |
308 | { | |
309 | unsigned long mask, pfn, pageno, start = 0; | |
310 | struct cma *cma = dev_get_cma_area(dev); | |
bdd43cb3 | 311 | struct page *page = NULL; |
c64be2bb MS |
312 | int ret; |
313 | ||
314 | if (!cma || !cma->count) | |
315 | return NULL; | |
316 | ||
317 | if (align > CONFIG_CMA_ALIGNMENT) | |
318 | align = CONFIG_CMA_ALIGNMENT; | |
319 | ||
320 | pr_debug("%s(cma %p, count %d, align %d)\n", __func__, (void *)cma, | |
321 | count, align); | |
322 | ||
323 | if (!count) | |
324 | return NULL; | |
325 | ||
326 | mask = (1 << align) - 1; | |
327 | ||
328 | mutex_lock(&cma_mutex); | |
329 | ||
330 | for (;;) { | |
331 | pageno = bitmap_find_next_zero_area(cma->bitmap, cma->count, | |
332 | start, count, mask); | |
bdd43cb3 MN |
333 | if (pageno >= cma->count) |
334 | break; | |
c64be2bb MS |
335 | |
336 | pfn = cma->base_pfn + pageno; | |
337 | ret = alloc_contig_range(pfn, pfn + count, MIGRATE_CMA); | |
338 | if (ret == 0) { | |
339 | bitmap_set(cma->bitmap, pageno, count); | |
bdd43cb3 | 340 | page = pfn_to_page(pfn); |
c64be2bb MS |
341 | break; |
342 | } else if (ret != -EBUSY) { | |
bdd43cb3 | 343 | break; |
c64be2bb MS |
344 | } |
345 | pr_debug("%s(): memory range at %p is busy, retrying\n", | |
346 | __func__, pfn_to_page(pfn)); | |
347 | /* try again with a bit different memory target */ | |
348 | start = pageno + mask + 1; | |
349 | } | |
350 | ||
351 | mutex_unlock(&cma_mutex); | |
bdd43cb3 MN |
352 | pr_debug("%s(): returned %p\n", __func__, page); |
353 | return page; | |
c64be2bb MS |
354 | } |
355 | ||
356 | /** | |
357 | * dma_release_from_contiguous() - release allocated pages | |
358 | * @dev: Pointer to device for which the pages were allocated. | |
359 | * @pages: Allocated pages. | |
360 | * @count: Number of allocated pages. | |
361 | * | |
362 | * This function releases memory allocated by dma_alloc_from_contiguous(). | |
363 | * It returns false when provided pages do not belong to contiguous area and | |
364 | * true otherwise. | |
365 | */ | |
366 | bool dma_release_from_contiguous(struct device *dev, struct page *pages, | |
367 | int count) | |
368 | { | |
369 | struct cma *cma = dev_get_cma_area(dev); | |
370 | unsigned long pfn; | |
371 | ||
372 | if (!cma || !pages) | |
373 | return false; | |
374 | ||
375 | pr_debug("%s(page %p)\n", __func__, (void *)pages); | |
376 | ||
377 | pfn = page_to_pfn(pages); | |
378 | ||
379 | if (pfn < cma->base_pfn || pfn >= cma->base_pfn + cma->count) | |
380 | return false; | |
381 | ||
382 | VM_BUG_ON(pfn + count > cma->base_pfn + cma->count); | |
383 | ||
384 | mutex_lock(&cma_mutex); | |
385 | bitmap_clear(cma->bitmap, pfn - cma->base_pfn, count); | |
386 | free_contig_range(pfn, count); | |
387 | mutex_unlock(&cma_mutex); | |
388 | ||
389 | return true; | |
390 | } |