Blame view
Documentation/DMA-API.txt
27.6 KB
03158a70a
|
1 2 3 |
============================================ Dynamic DMA mapping using the generic device ============================================ |
1da177e4c
|
4 |
|
03158a70a
|
5 |
:Author: James E.J. Bottomley <James.Bottomley@HansenPartnership.com> |
1da177e4c
|
6 7 |
This document describes the DMA API. For a more gentle introduction |
77f2ea2f8
|
8 |
of the API (and actual examples), see Documentation/DMA-API-HOWTO.txt. |
1da177e4c
|
9 |
|
77f2ea2f8
|
10 11 12 13 14 |
This API is split into two pieces. Part I describes the basic API. Part II describes extensions for supporting non-consistent memory machines. Unless you know that your driver absolutely has to support non-consistent platforms (this is usually only legacy platforms) you should only use the API described in part I. |
1da177e4c
|
15 |
|
03158a70a
|
16 17 |
Part I - dma_API ---------------- |
1da177e4c
|
18 |
|
03158a70a
|
19 |
To get the dma_API, you must #include <linux/dma-mapping.h>. This |
77f2ea2f8
|
20 |
provides dma_addr_t and the interfaces described below. |
1da177e4c
|
21 |
|
3a9ad0b4f
|
22 23 24 25 |
A dma_addr_t can hold any valid DMA address for the platform. It can be given to a device to use as a DMA source or target. A CPU cannot reference a dma_addr_t directly because there may be translation between its physical address space and the DMA address space. |
1da177e4c
|
26 |
|
77f2ea2f8
|
27 |
Part Ia - Using large DMA-coherent buffers |
1da177e4c
|
28 |
------------------------------------------ |
03158a70a
|
29 30 31 32 33 |
:: void * dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flag) |
1da177e4c
|
34 35 36 |
Consistent memory is memory for which a write by either the device or the processor can immediately be read by the processor or device |
21440d313
|
37 38 39 |
without having to worry about caching effects. (You may however need to make sure to flush the processor's write buffers before telling devices to read that memory.) |
1da177e4c
|
40 41 |
This routine allocates a region of <size> bytes of consistent memory. |
1da177e4c
|
42 |
|
77f2ea2f8
|
43 |
It returns a pointer to the allocated region (in the processor's virtual |
1da177e4c
|
44 |
address space) or NULL if the allocation failed. |
77f2ea2f8
|
45 |
It also returns a <dma_handle> which may be cast to an unsigned integer the |
3a9ad0b4f
|
46 |
same width as the bus and given to the device as the DMA address base of |
77f2ea2f8
|
47 |
the region. |
1da177e4c
|
48 49 50 51 |
Note: consistent memory can be expensive on some platforms, and the minimum allocation length may be as big as a page, so you should consolidate your requests for consistent memory as much as possible. The simplest way to do that is to use the dma_pool calls (see below). |
77f2ea2f8
|
52 |
The flag parameter (dma_alloc_coherent() only) allows the caller to |
03158a70a
|
53 |
specify the ``GFP_`` flags (see kmalloc()) for the allocation (the |
a12e2c6cd
|
54 |
implementation may choose to ignore flags that affect the location of |
f5a69f4c1
|
55 |
the returned memory, like GFP_DMA). |
1da177e4c
|
56 |
|
03158a70a
|
57 58 59 60 61 |
:: void * dma_zalloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flag) |
842fa69f3
|
62 63 64 |
Wraps dma_alloc_coherent() and also zeroes the returned memory if the allocation attempt succeeded. |
03158a70a
|
65 66 67 68 69 |
:: void dma_free_coherent(struct device *dev, size_t size, void *cpu_addr, dma_addr_t dma_handle) |
1da177e4c
|
70 |
|
77f2ea2f8
|
71 72 73 74 |
Free a region of consistent memory you previously allocated. dev, size and dma_handle must all be the same as those passed into dma_alloc_coherent(). cpu_addr must be the virtual address returned by the dma_alloc_coherent(). |
1da177e4c
|
75 |
|
aa24886e3
|
76 77 |
Note that unlike their sibling allocation calls, these routines may only be called with IRQs enabled. |
1da177e4c
|
78 |
|
77f2ea2f8
|
79 |
Part Ib - Using small DMA-coherent buffers |
1da177e4c
|
80 |
------------------------------------------ |
03158a70a
|
81 |
To get this part of the dma_API, you must #include <linux/dmapool.h> |
1da177e4c
|
82 |
|
77f2ea2f8
|
83 |
Many drivers need lots of small DMA-coherent memory regions for DMA |
1da177e4c
|
84 85 |
descriptors or I/O buffers. Rather than allocating in units of a page or more using dma_alloc_coherent(), you can use DMA pools. These work |
77f2ea2f8
|
86 |
much like a struct kmem_cache, except that they use the DMA-coherent allocator, |
1da177e4c
|
87 |
not __get_free_pages(). Also, they understand common hardware constraints |
a12e2c6cd
|
88 |
for alignment, like queue heads needing to be aligned on N-byte boundaries. |
1da177e4c
|
89 |
|
03158a70a
|
90 |
:: |
1da177e4c
|
91 92 93 |
struct dma_pool * dma_pool_create(const char *name, struct device *dev, size_t size, size_t align, size_t alloc); |
77f2ea2f8
|
94 |
dma_pool_create() initializes a pool of DMA-coherent buffers |
1da177e4c
|
95 96 |
for use with a given device. It must be called in a context which can sleep. |
e18b890bb
|
97 |
The "name" is for diagnostics (like a struct kmem_cache name); dev and size |
1da177e4c
|
98 99 100 101 102 |
are like what you'd pass to dma_alloc_coherent(). The device's hardware alignment requirement for this type of data is "align" (which is expressed in bytes, and must be a power of two). If your device has no boundary crossing restrictions, pass 0 for alloc; passing 4096 says memory allocated from this pool must not cross 4KByte boundaries. |
03158a70a
|
103 |
:: |
1da177e4c
|
104 |
|
03158a70a
|
105 106 107 |
void * dma_pool_zalloc(struct dma_pool *pool, gfp_t mem_flags, dma_addr_t *handle) |
ad82362b2
|
108 109 110 |
Wraps dma_pool_alloc() and also zeroes the returned memory if the allocation attempt succeeded. |
03158a70a
|
111 112 113 114 115 |
:: void * dma_pool_alloc(struct dma_pool *pool, gfp_t gfp_flags, dma_addr_t *dma_handle); |
1da177e4c
|
116 |
|
77f2ea2f8
|
117 118 119 120 121 |
This allocates memory from the pool; the returned memory will meet the size and alignment requirements specified at creation time. Pass GFP_ATOMIC to prevent blocking, or if it's permitted (not in_interrupt, not holding SMP locks), pass GFP_KERNEL to allow blocking. Like dma_alloc_coherent(), this returns two values: an |
f311a724a
|
122 |
address usable by the CPU, and the DMA address usable by the pool's |
77f2ea2f8
|
123 |
device. |
1da177e4c
|
124 |
|
03158a70a
|
125 |
:: |
1da177e4c
|
126 |
|
03158a70a
|
127 128 129 |
void dma_pool_free(struct dma_pool *pool, void *vaddr, dma_addr_t addr); |
1da177e4c
|
130 |
|
1da177e4c
|
131 |
This puts memory back into the pool. The pool is what was passed to |
f311a724a
|
132 |
dma_pool_alloc(); the CPU (vaddr) and DMA addresses are what |
1da177e4c
|
133 |
were returned when that routine allocated the memory being freed. |
03158a70a
|
134 |
:: |
1da177e4c
|
135 |
|
03158a70a
|
136 137 |
void dma_pool_destroy(struct dma_pool *pool); |
1da177e4c
|
138 |
|
77f2ea2f8
|
139 |
dma_pool_destroy() frees the resources of the pool. It must be |
1da177e4c
|
140 141 142 143 144 145 |
called in a context which can sleep. Make sure you've freed all allocated memory back to the pool before you destroy it. Part Ic - DMA addressing limitations ------------------------------------ |
03158a70a
|
146 147 148 149 |
:: int dma_set_mask_and_coherent(struct device *dev, u64 mask) |
4aa806b77
|
150 151 152 153 154 |
Checks to see if the mask is possible and updates the device streaming and coherent DMA mask parameters if it is. Returns: 0 if successful and a negative error if not. |
03158a70a
|
155 156 157 158 |
:: int dma_set_mask(struct device *dev, u64 mask) |
1da177e4c
|
159 160 161 162 163 |
Checks to see if the mask is possible and updates the device parameters if it is. Returns: 0 if successful and a negative error if not. |
03158a70a
|
164 165 166 167 |
:: int dma_set_coherent_mask(struct device *dev, u64 mask) |
6a1961f49
|
168 169 170 171 172 |
Checks to see if the mask is possible and updates the device parameters if it is. Returns: 0 if successful and a negative error if not. |
03158a70a
|
173 174 175 176 |
:: u64 dma_get_required_mask(struct device *dev) |
1da177e4c
|
177 |
|
175add198
|
178 179 |
This API returns the mask that the platform requires to operate efficiently. Usually this means the returned mask |
1da177e4c
|
180 181 182 183 184 |
is the minimum required to cover all of memory. Examining the required mask gives drivers with variable descriptor sizes the opportunity to use smaller descriptors as necessary. Requesting the required mask does not alter the current mask. If you |
175add198
|
185 186 |
wish to take advantage of it, you should issue a dma_set_mask() call to set the mask to the value returned. |
1da177e4c
|
187 188 189 190 |
Part Id - Streaming DMA mappings -------------------------------- |
03158a70a
|
191 192 193 194 195 |
:: dma_addr_t dma_map_single(struct device *dev, void *cpu_addr, size_t size, enum dma_data_direction direction) |
1da177e4c
|
196 197 |
Maps a piece of processor virtual memory so it can be accessed by the |
3a9ad0b4f
|
198 |
device and returns the DMA address of the memory. |
1da177e4c
|
199 |
|
77f2ea2f8
|
200 |
The direction for both APIs may be converted freely by casting. |
03158a70a
|
201 |
However the dma_API uses a strongly typed enumerator for its |
1da177e4c
|
202 |
direction: |
03158a70a
|
203 |
======================= ============================================= |
f5a69f4c1
|
204 205 206 207 |
DMA_NONE no direction (used for debugging) DMA_TO_DEVICE data is going from the memory to the device DMA_FROM_DEVICE data is coming from the device to the memory DMA_BIDIRECTIONAL direction isn't known |
03158a70a
|
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 |
======================= ============================================= .. note:: Not all memory regions in a machine can be mapped by this API. Further, contiguous kernel virtual space may not be contiguous as physical memory. Since this API does not provide any scatter/gather capability, it will fail if the user tries to map a non-physically contiguous piece of memory. For this reason, memory to be mapped by this API should be obtained from sources which guarantee it to be physically contiguous (like kmalloc). Further, the DMA address of the memory must be within the dma_mask of the device (the dma_mask is a bit mask of the addressable region for the device, i.e., if the DMA address of the memory ANDed with the dma_mask is still equal to the DMA address, then the device can perform DMA to the memory). To ensure that the memory allocated by kmalloc is within the dma_mask, the driver may specify various platform-dependent flags to restrict the DMA address range of the allocation (e.g., on x86, GFP_DMA guarantees to be within the first 16MB of available DMA addresses, as required by ISA devices). Note also that the above constraints on physical contiguity and dma_mask may not apply if the platform has an IOMMU (a device which maps an I/O DMA address to a physical memory address). However, to be portable, device driver writers may *not* assume that such an IOMMU exists. .. warning:: Memory coherency operates at a granularity called the cache line width. In order for memory mapped by this API to operate correctly, the mapped region must begin exactly on a cache line boundary and end exactly on one (to prevent two separately mapped regions from sharing a single cache line). Since the cache line size may not be known at compile time, the API will not enforce this requirement. Therefore, it is recommended that driver writers who don't take special care to determine the cache line size at run time only map virtual regions that begin and end on page boundaries (which are guaranteed also to be cache line boundaries). DMA_TO_DEVICE synchronisation must be done after the last modification of the memory region by the software and before it is handed off to the device. Once this primitive is used, memory covered by this primitive should be treated as read-only by the device. If the device may write to it at any point, it should be DMA_BIDIRECTIONAL (see below). DMA_FROM_DEVICE synchronisation must be done before the driver accesses data that may be changed by the device. This memory should be treated as read-only by the driver. If the driver needs to write to it at any point, it should be DMA_BIDIRECTIONAL (see below). DMA_BIDIRECTIONAL requires special handling: it means that the driver isn't sure if the memory was modified before being handed off to the device and also isn't sure if the device will also modify it. Thus, you must always sync bidirectional memory twice: once before the memory is handed off to the device (to make sure all memory changes are flushed from the processor) and once before the data may be accessed after being used by the device (to make sure any processor cache lines are updated with data that the device may have changed). :: |
1da177e4c
|
272 |
|
03158a70a
|
273 274 275 |
void dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size, enum dma_data_direction direction) |
1da177e4c
|
276 277 278 279 |
Unmaps the region previously mapped. All the parameters passed in must be identical to those passed in (and returned) by the mapping API. |
03158a70a
|
280 281 282 283 284 285 286 287 288 289 |
:: dma_addr_t dma_map_page(struct device *dev, struct page *page, unsigned long offset, size_t size, enum dma_data_direction direction) void dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size, enum dma_data_direction direction) |
1da177e4c
|
290 291 292 293 294 295 |
API for mapping and unmapping for pages. All the notes and warnings for the other mapping APIs apply here. Also, although the <offset> and <size> parameters are provided to do partial page mapping, it is recommended that you never use these unless you really know what the cache width is. |
03158a70a
|
296 |
:: |
6f3d87968
|
297 |
|
03158a70a
|
298 299 300 301 302 303 304 |
dma_addr_t dma_map_resource(struct device *dev, phys_addr_t phys_addr, size_t size, enum dma_data_direction dir, unsigned long attrs) void dma_unmap_resource(struct device *dev, dma_addr_t addr, size_t size, enum dma_data_direction dir, unsigned long attrs) |
6f3d87968
|
305 306 307 308 |
API for mapping and unmapping for MMIO resources. All the notes and warnings for the other mapping APIs apply here. The API should only be used to map device MMIO resources, mapping of RAM is not permitted. |
03158a70a
|
309 310 311 312 |
:: int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) |
1da177e4c
|
313 |
|
6f3d87968
|
314 315 316 317 318 |
In some circumstances dma_map_single(), dma_map_page() and dma_map_resource() will fail to create a mapping. A driver can check for these errors by testing the returned DMA address with dma_mapping_error(). A non-zero return value means the mapping could not be created and the driver should take appropriate action (e.g. reduce current DMA mapping usage or delay and try again later). |
1da177e4c
|
319 |
|
03158a70a
|
320 |
:: |
21440d313
|
321 322 |
int dma_map_sg(struct device *dev, struct scatterlist *sg, |
03158a70a
|
323 |
int nents, enum dma_data_direction direction) |
1da177e4c
|
324 |
|
3a9ad0b4f
|
325 |
Returns: the number of DMA address segments mapped (this may be shorter |
1d678f365
|
326 327 328 |
than <nents> passed in if some elements of the scatter/gather list are physically or virtually adjacent and an IOMMU maps them with a single entry). |
1da177e4c
|
329 330 331 |
Please note that the sg cannot be mapped again if it has been mapped once. The mapping process is allowed to destroy information in the sg. |
77f2ea2f8
|
332 |
As with the other mapping interfaces, dma_map_sg() can fail. When it |
1da177e4c
|
333 334 335 336 |
does, 0 is returned and a driver must take appropriate action. It is critical that the driver do something, in the case of a block driver aborting the request or even oopsing is better than doing nothing and corrupting the filesystem. |
03158a70a
|
337 |
With scatterlists, you use the resulting mapping like this:: |
21440d313
|
338 339 340 |
int i, count = dma_map_sg(dev, sglist, nents, direction); struct scatterlist *sg; |
79eb01457
|
341 |
for_each_sg(sglist, sg, count, i) { |
21440d313
|
342 343 344 345 346 347 348 349 350 351 352 353 354 355 |
hw_address[i] = sg_dma_address(sg); hw_len[i] = sg_dma_len(sg); } where nents is the number of entries in the sglist. The implementation is free to merge several consecutive sglist entries into one (e.g. with an IOMMU, or if several pages just happen to be physically contiguous) and returns the actual number of sg entries it mapped them to. On failure 0, is returned. Then you should loop count times (note: this can be less than nents times) and use sg_dma_address() and sg_dma_len() macros where you previously accessed sg->address and sg->length as shown above. |
03158a70a
|
356 |
:: |
21440d313
|
357 358 |
void dma_unmap_sg(struct device *dev, struct scatterlist *sg, |
03158a70a
|
359 |
int nents, enum dma_data_direction direction) |
1da177e4c
|
360 |
|
a12e2c6cd
|
361 |
Unmap the previously mapped scatter/gather list. All the parameters |
1da177e4c
|
362 363 364 365 |
must be the same as those and passed in to the scatter/gather mapping API. Note: <nents> must be the number you passed in, *not* the number of |
3a9ad0b4f
|
366 |
DMA address entries returned. |
1da177e4c
|
367 |
|
03158a70a
|
368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 |
:: void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction) void dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction) void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nents, enum dma_data_direction direction) void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nents, enum dma_data_direction direction) |
9705ef7ec
|
389 |
|
f311a724a
|
390 |
Synchronise a single contiguous or scatter/gather mapping for the CPU |
9705ef7ec
|
391 392 393 394 |
and device. With the sync_sg API, all the parameters must be the same as those passed into the single mapping API. With the sync_single API, you can use dma_handle and size parameters that aren't identical to those passed into the single mapping API to do a partial sync. |
9705ef7ec
|
395 |
|
03158a70a
|
396 397 398 399 400 401 402 403 404 405 |
.. note:: You must do this: - Before reading values that have been written by DMA from the device (use the DMA_FROM_DEVICE direction) - After writing values that will be written to the device using DMA (use the DMA_TO_DEVICE) direction - before *and* after handing memory to the device if the memory is DMA_BIDIRECTIONAL |
9705ef7ec
|
406 407 |
See also dma_map_single(). |
03158a70a
|
408 409 410 411 412 413 |
:: dma_addr_t dma_map_single_attrs(struct device *dev, void *cpu_addr, size_t size, enum dma_data_direction dir, unsigned long attrs) |
a75b0a2f6
|
414 |
|
03158a70a
|
415 416 417 418 |
void dma_unmap_single_attrs(struct device *dev, dma_addr_t dma_addr, size_t size, enum dma_data_direction dir, unsigned long attrs) |
a75b0a2f6
|
419 |
|
03158a70a
|
420 421 422 423 |
int dma_map_sg_attrs(struct device *dev, struct scatterlist *sgl, int nents, enum dma_data_direction dir, unsigned long attrs) |
a75b0a2f6
|
424 |
|
03158a70a
|
425 426 427 428 |
void dma_unmap_sg_attrs(struct device *dev, struct scatterlist *sgl, int nents, enum dma_data_direction dir, unsigned long attrs) |
a75b0a2f6
|
429 430 431 |
The four functions above are just like the counterpart functions without the _attrs suffixes, except that they pass an optional |
00085f1ef
|
432 |
dma_attrs. |
a75b0a2f6
|
433 |
|
77f2ea2f8
|
434 |
The interpretation of DMA attributes is architecture-specific, and |
a75b0a2f6
|
435 |
each attribute should be documented in Documentation/DMA-attributes.txt. |
00085f1ef
|
436 437 |
If dma_attrs are 0, the semantics of each of these functions is identical to those of the corresponding function |
a75b0a2f6
|
438 439 |
without the _attrs suffix. As a result dma_map_single_attrs() can generally replace dma_map_single(), etc. |
03158a70a
|
440 |
As an example of the use of the ``*_attrs`` functions, here's how |
a75b0a2f6
|
441 |
you could pass an attribute DMA_ATTR_FOO when mapping memory |
03158a70a
|
442 |
for DMA:: |
a75b0a2f6
|
443 |
|
03158a70a
|
444 445 446 447 |
#include <linux/dma-mapping.h> /* DMA_ATTR_FOO should be defined in linux/dma-mapping.h and * documented in Documentation/DMA-attributes.txt */ ... |
a75b0a2f6
|
448 |
|
03158a70a
|
449 450 451 452 453 |
unsigned long attr; attr |= DMA_ATTR_FOO; .... n = dma_map_sg_attrs(dev, sg, nents, DMA_TO_DEVICE, attr); .... |
a75b0a2f6
|
454 455 456 |
Architectures that care about DMA_ATTR_FOO would check for its presence in their implementations of the mapping and unmapping |
03158a70a
|
457 458 459 460 461 462 463 464 465 466 467 |
routines, e.g.::: void whizco_dma_map_sg_attrs(struct device *dev, dma_addr_t dma_addr, size_t size, enum dma_data_direction dir, unsigned long attrs) { .... if (attrs & DMA_ATTR_FOO) /* twizzle the frobnozzle */ .... } |
a75b0a2f6
|
468 |
|
1da177e4c
|
469 |
|
03158a70a
|
470 471 |
Part II - Advanced dma usage ---------------------------- |
1da177e4c
|
472 |
|
f5a69f4c1
|
473 474 475 |
Warning: These pieces of the DMA API should not be used in the majority of cases, since they cater for unlikely corner cases that don't belong in usual drivers. |
1da177e4c
|
476 477 478 479 |
If you don't understand how cache line coherency works between a processor and an I/O device, you should not be using this part of the API at all. |
03158a70a
|
480 481 482 |
:: void * |
2fd523c57
|
483 484 |
dma_alloc_attrs(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flag, unsigned long attrs) |
1da177e4c
|
485 |
|
2fd523c57
|
486 487 488 489 490 491 |
Identical to dma_alloc_coherent() except that when the DMA_ATTR_NON_CONSISTENT flags is passed in the attrs argument, the platform will choose to return either consistent or non-consistent memory as it sees fit. By using this API, you are guaranteeing to the platform that you have all the correct and necessary sync points for this memory in the driver should it choose to return non-consistent memory. |
1da177e4c
|
492 493 494 495 496 |
Note: where the platform can return consistent memory, it will guarantee that the sync points become nops. Warning: Handling non-consistent memory is a real pain. You should |
77f2ea2f8
|
497 |
only use this API if you positively know your driver will be |
1da177e4c
|
498 499 |
required to work on one of the rare (usually non-PCI) architectures that simply cannot make consistent memory. |
03158a70a
|
500 501 502 |
:: void |
2fd523c57
|
503 504 |
dma_free_attrs(struct device *dev, size_t size, void *cpu_addr, dma_addr_t dma_handle, unsigned long attrs) |
1da177e4c
|
505 |
|
2fd523c57
|
506 507 508 509 |
Free memory allocated by the dma_alloc_attrs(). All parameters common parameters must identical to those otherwise passed to dma_fre_coherent, and the attrs argument must be identical to the attrs passed to dma_alloc_attrs(). |
1da177e4c
|
510 |
|
03158a70a
|
511 512 513 514 |
:: int dma_get_cache_alignment(void) |
1da177e4c
|
515 |
|
a12e2c6cd
|
516 |
Returns the processor cache alignment. This is the absolute minimum |
1da177e4c
|
517 518 |
alignment *and* width that you must observe when either mapping memory or doing partial flushes. |
03158a70a
|
519 |
.. note:: |
1da177e4c
|
520 |
|
03158a70a
|
521 522 523 524 525 526 527 528 529 530 |
This API may return a number *larger* than the actual cache line, but it will guarantee that one or more cache lines fit exactly into the width returned by this call. It will also always be a power of two for easy alignment. :: void dma_cache_sync(struct device *dev, void *vaddr, size_t size, enum dma_data_direction direction) |
1da177e4c
|
531 |
|
2fd523c57
|
532 533 |
Do a partial sync of memory that was allocated by dma_alloc_attrs() with the DMA_ATTR_NON_CONSISTENT flag starting at virtual address vaddr and |
1da177e4c
|
534 535 |
continuing on for size. Again, you *must* observe the cache line boundaries when doing this. |
03158a70a
|
536 537 538 539 540 541 |
:: int dma_declare_coherent_memory(struct device *dev, phys_addr_t phys_addr, dma_addr_t device_addr, size_t size, int flags) |
1da177e4c
|
542 |
|
77f2ea2f8
|
543 |
Declare region of memory to be handed out by dma_alloc_coherent() when |
1da177e4c
|
544 |
it's asked for coherent memory for this device. |
f311a724a
|
545 546 |
phys_addr is the CPU physical address to which the memory is currently assigned (this will be ioremapped so the CPU can access the region). |
1da177e4c
|
547 |
|
3a9ad0b4f
|
548 |
device_addr is the DMA address the device needs to be programmed |
88a984ba0
|
549 |
with to actually address this memory (this will be handed out as the |
a12e2c6cd
|
550 |
dma_addr_t in dma_alloc_coherent()). |
1da177e4c
|
551 552 |
size is the size of the area (must be multiples of PAGE_SIZE). |
77f2ea2f8
|
553 |
flags can be ORed together and are: |
1da177e4c
|
554 |
|
03158a70a
|
555 556 557 |
- DMA_MEMORY_EXCLUSIVE - only allocate memory from the declared regions. Do not allow dma_alloc_coherent() to fall back to system memory when it's out of memory in the declared region. |
1da177e4c
|
558 |
|
2436bdcda
|
559 |
As a simplification for the platforms, only *one* such region of |
1da177e4c
|
560 561 562 563 564 |
memory may be declared per device. For reasons of efficiency, most platforms choose to track the declared region only at the granularity of a page. For smaller allocations, you should use the dma_pool() API. |
03158a70a
|
565 566 567 568 |
:: void dma_release_declared_memory(struct device *dev) |
1da177e4c
|
569 570 571 572 |
Remove the memory region previously declared from the system. This API performs *no* in-use checking for this region and will return unconditionally having removed all the required structures. It is the |
a12e2c6cd
|
573 |
driver's job to ensure that no parts of this memory region are |
1da177e4c
|
574 |
currently in use. |
03158a70a
|
575 576 577 578 579 |
:: void * dma_mark_declared_memory_occupied(struct device *dev, dma_addr_t device_addr, size_t size) |
1da177e4c
|
580 581 582 |
This is used to occupy specific regions of the declared space (dma_alloc_coherent() will hand out the first free region it finds). |
a12e2c6cd
|
583 |
device_addr is the *device* address of the region requested. |
1da177e4c
|
584 |
|
a12e2c6cd
|
585 |
size is the size (and should be a page-sized multiple). |
1da177e4c
|
586 587 588 589 |
The return value will be either a pointer to the processor virtual address of the memory, or an error (via PTR_ERR()) if any part of the region is occupied. |
187f9c3f0
|
590 591 592 |
Part III - Debug drivers use of the DMA-API ------------------------------------------- |
77f2ea2f8
|
593 |
The DMA-API as described above has some constraints. DMA addresses must be |
187f9c3f0
|
594 595 596 597 598 599 600 601 602 603 604 605 606 607 |
released with the corresponding function with the same size for example. With the advent of hardware IOMMUs it becomes more and more important that drivers do not violate those constraints. In the worst case such a violation can result in data corruption up to destroyed filesystems. To debug drivers and find bugs in the usage of the DMA-API checking code can be compiled into the kernel which will tell the developer about those violations. If your architecture supports it you can select the "Enable debugging of DMA-API usage" option in your kernel configuration. Enabling this option has a performance impact. Do not enable it in production kernels. If you boot the resulting kernel will contain code which does some bookkeeping about what DMA memory was allocated for which device. If this code detects an error it prints a warning message with some details into your kernel log. An |
03158a70a
|
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 |
example warning message may look like this:: WARNING: at /data2/repos/linux-2.6-iommu/lib/dma-debug.c:448 check_unmap+0x203/0x490() Hardware name: forcedeth 0000:00:08.0: DMA-API: device driver frees DMA memory with wrong function [device address=0x00000000640444be] [size=66 bytes] [mapped as single] [unmapped as page] Modules linked in: nfsd exportfs bridge stp llc r8169 Pid: 0, comm: swapper Tainted: G W 2.6.28-dmatest-09289-g8bb99c0 #1 Call Trace: <IRQ> [<ffffffff80240b22>] warn_slowpath+0xf2/0x130 [<ffffffff80647b70>] _spin_unlock+0x10/0x30 [<ffffffff80537e75>] usb_hcd_link_urb_to_ep+0x75/0xc0 [<ffffffff80647c22>] _spin_unlock_irqrestore+0x12/0x40 [<ffffffff8055347f>] ohci_urb_enqueue+0x19f/0x7c0 [<ffffffff80252f96>] queue_work+0x56/0x60 [<ffffffff80237e10>] enqueue_task_fair+0x20/0x50 [<ffffffff80539279>] usb_hcd_submit_urb+0x379/0xbc0 [<ffffffff803b78c3>] cpumask_next_and+0x23/0x40 [<ffffffff80235177>] find_busiest_group+0x207/0x8a0 [<ffffffff8064784f>] _spin_lock_irqsave+0x1f/0x50 [<ffffffff803c7ea3>] check_unmap+0x203/0x490 [<ffffffff803c8259>] debug_dma_unmap_page+0x49/0x50 [<ffffffff80485f26>] nv_tx_done_optimized+0xc6/0x2c0 [<ffffffff80486c13>] nv_nic_irq_optimized+0x73/0x2b0 [<ffffffff8026df84>] handle_IRQ_event+0x34/0x70 [<ffffffff8026ffe9>] handle_edge_irq+0xc9/0x150 [<ffffffff8020e3ab>] do_IRQ+0xcb/0x1c0 [<ffffffff8020c093>] ret_from_intr+0x0/0xa <EOI> <4>---[ end trace f6435a98e2a38c0e ]--- |
187f9c3f0
|
639 640 641 642 643 644 645 646 647 648 649 650 |
The driver developer can find the driver and the device including a stacktrace of the DMA-API call which caused this warning. Per default only the first error will result in a warning message. All other errors will only silently counted. This limitation exist to prevent the code from flooding your kernel log. To support debugging a device driver this can be disabled via debugfs. See the debugfs interface documentation below for details. The debugfs directory for the DMA-API debugging code is called dma-api/. In this directory the following files can currently be found: |
03158a70a
|
651 652 |
=============================== =============================================== dma-api/all_errors This file contains a numeric value. If this |
187f9c3f0
|
653 654 |
value is not equal to zero the debugging code will print a warning for every error it finds |
19f594600
|
655 656 |
into the kernel log. Be careful with this option, as it can easily flood your logs. |
187f9c3f0
|
657 |
|
03158a70a
|
658 |
dma-api/disabled This read-only file contains the character 'Y' |
187f9c3f0
|
659 660 661 |
if the debugging code is disabled. This can happen when it runs out of memory or if it was disabled at boot time |
03158a70a
|
662 |
dma-api/error_count This file is read-only and shows the total |
187f9c3f0
|
663 |
numbers of errors found. |
03158a70a
|
664 |
dma-api/num_errors The number in this file shows how many |
187f9c3f0
|
665 666 667 668 |
warnings will be printed to the kernel log before it stops. This number is initialized to one at system boot and be set by writing into this file |
03158a70a
|
669 |
dma-api/min_free_entries This read-only file can be read to get the |
187f9c3f0
|
670 671 672 673 |
minimum number of free dma_debug_entries the allocator has ever seen. If this value goes down to zero the code will disable itself because it is not longer reliable. |
03158a70a
|
674 |
dma-api/num_free_entries The current number of free dma_debug_entries |
187f9c3f0
|
675 |
in the allocator. |
03158a70a
|
676 |
dma-api/driver-filter You can write a name of a driver into this file |
016ea6874
|
677 678 679 680 |
to limit the debug output to requests from that particular driver. Write an empty string to that file to disable the filter and see all errors again. |
03158a70a
|
681 |
=============================== =============================================== |
016ea6874
|
682 |
|
187f9c3f0
|
683 684 685 686 687 |
If you have this code compiled into your kernel it will be enabled by default. If you want to boot without the bookkeeping anyway you can provide 'dma_debug=off' as a boot parameter. This will disable DMA-API debugging. Notice that you can not enable it again at runtime. You have to reboot to do so. |
016ea6874
|
688 689 690 691 |
If you want to see debug messages only for a special device driver you can specify the dma_debug_driver=<drivername> parameter. This will enable the driver filter at boot time. The debug code will only print errors for that driver afterwards. This filter can be disabled or changed later using debugfs. |
187f9c3f0
|
692 693 694 695 696 |
When the code disables itself at runtime this is most likely because it ran out of dma_debug_entries. These entries are preallocated at boot. The number of preallocated entries is defined per architecture. If it is too low for you boot with 'dma_debug_entries=<your_desired_number>' to overwrite the architectural default. |
6c9c6d630
|
697 |
|
03158a70a
|
698 699 700 701 |
:: void debug_dma_mapping_error(struct device *dev, dma_addr_t dma_addr); |
6c9c6d630
|
702 703 |
dma-debug interface debug_dma_mapping_error() to debug drivers that fail |
77f2ea2f8
|
704 |
to check DMA mapping errors on addresses returned by dma_map_single() and |
6c9c6d630
|
705 706 707 708 709 |
dma_map_page() interfaces. This interface clears a flag set by debug_dma_map_page() to indicate that dma_mapping_error() has been called by the driver. When driver does unmap, debug_dma_unmap() checks the flag and if this flag is still set, prints warning message that includes call trace that leads up to the unmap. This interface can be called from dma_mapping_error() |
77f2ea2f8
|
710 |
routines to enable DMA mapping error check debugging. |