Blame view
Documentation/DMA-API.txt
27.5 KB
1da177e4c Linux-2.6.12-rc2 |
1 2 3 4 5 6 |
Dynamic DMA mapping using the generic device ============================================ James E.J. Bottomley <James.Bottomley@HansenPartnership.com> This document describes the DMA API. For a more gentle introduction |
77f2ea2f8 DMA-API: Clarify ... |
7 |
of the API (and actual examples), see Documentation/DMA-API-HOWTO.txt. |
1da177e4c Linux-2.6.12-rc2 |
8 |
|
77f2ea2f8 DMA-API: Clarify ... |
9 10 11 12 13 |
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 Linux-2.6.12-rc2 |
14 |
|
f5a69f4c1 Documentation: re... |
15 |
Part I - dma_ API |
1da177e4c Linux-2.6.12-rc2 |
16 |
------------------------------------- |
77f2ea2f8 DMA-API: Clarify ... |
17 18 |
To get the dma_ API, you must #include <linux/dma-mapping.h>. This provides dma_addr_t and the interfaces described below. |
1da177e4c Linux-2.6.12-rc2 |
19 |
|
3a9ad0b4f PCI: Add pci_bus_... |
20 21 22 23 |
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 Linux-2.6.12-rc2 |
24 |
|
77f2ea2f8 DMA-API: Clarify ... |
25 |
Part Ia - Using large DMA-coherent buffers |
1da177e4c Linux-2.6.12-rc2 |
26 27 28 29 |
------------------------------------------ void * dma_alloc_coherent(struct device *dev, size_t size, |
a12e2c6cd Doc: DMA-API update |
30 |
dma_addr_t *dma_handle, gfp_t flag) |
1da177e4c Linux-2.6.12-rc2 |
31 32 33 |
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 [PATCH] dma doc u... |
34 35 36 |
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 Linux-2.6.12-rc2 |
37 38 |
This routine allocates a region of <size> bytes of consistent memory. |
1da177e4c Linux-2.6.12-rc2 |
39 |
|
77f2ea2f8 DMA-API: Clarify ... |
40 |
It returns a pointer to the allocated region (in the processor's virtual |
1da177e4c Linux-2.6.12-rc2 |
41 |
address space) or NULL if the allocation failed. |
77f2ea2f8 DMA-API: Clarify ... |
42 |
It also returns a <dma_handle> which may be cast to an unsigned integer the |
3a9ad0b4f PCI: Add pci_bus_... |
43 |
same width as the bus and given to the device as the DMA address base of |
77f2ea2f8 DMA-API: Clarify ... |
44 |
the region. |
1da177e4c Linux-2.6.12-rc2 |
45 46 47 48 |
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 DMA-API: Clarify ... |
49 50 |
The flag parameter (dma_alloc_coherent() only) allows the caller to specify the GFP_ flags (see kmalloc()) for the allocation (the |
a12e2c6cd Doc: DMA-API update |
51 |
implementation may choose to ignore flags that affect the location of |
f5a69f4c1 Documentation: re... |
52 |
the returned memory, like GFP_DMA). |
1da177e4c Linux-2.6.12-rc2 |
53 |
|
842fa69f3 include/linux/dma... |
54 55 56 57 58 59 |
void * dma_zalloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flag) Wraps dma_alloc_coherent() and also zeroes the returned memory if the allocation attempt succeeded. |
1da177e4c Linux-2.6.12-rc2 |
60 |
void |
a12e2c6cd Doc: DMA-API update |
61 |
dma_free_coherent(struct device *dev, size_t size, void *cpu_addr, |
1da177e4c Linux-2.6.12-rc2 |
62 |
dma_addr_t dma_handle) |
1da177e4c Linux-2.6.12-rc2 |
63 |
|
77f2ea2f8 DMA-API: Clarify ... |
64 65 66 67 |
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 Linux-2.6.12-rc2 |
68 |
|
aa24886e3 dma_free_coherent... |
69 70 |
Note that unlike their sibling allocation calls, these routines may only be called with IRQs enabled. |
1da177e4c Linux-2.6.12-rc2 |
71 |
|
77f2ea2f8 DMA-API: Clarify ... |
72 |
Part Ib - Using small DMA-coherent buffers |
1da177e4c Linux-2.6.12-rc2 |
73 74 75 |
------------------------------------------ To get this part of the dma_ API, you must #include <linux/dmapool.h> |
77f2ea2f8 DMA-API: Clarify ... |
76 |
Many drivers need lots of small DMA-coherent memory regions for DMA |
1da177e4c Linux-2.6.12-rc2 |
77 78 |
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 DMA-API: Clarify ... |
79 |
much like a struct kmem_cache, except that they use the DMA-coherent allocator, |
1da177e4c Linux-2.6.12-rc2 |
80 |
not __get_free_pages(). Also, they understand common hardware constraints |
a12e2c6cd Doc: DMA-API update |
81 |
for alignment, like queue heads needing to be aligned on N-byte boundaries. |
1da177e4c Linux-2.6.12-rc2 |
82 83 84 85 86 |
struct dma_pool * dma_pool_create(const char *name, struct device *dev, size_t size, size_t align, size_t alloc); |
77f2ea2f8 DMA-API: Clarify ... |
87 |
dma_pool_create() initializes a pool of DMA-coherent buffers |
1da177e4c Linux-2.6.12-rc2 |
88 89 |
for use with a given device. It must be called in a context which can sleep. |
e18b890bb [PATCH] slab: rem... |
90 |
The "name" is for diagnostics (like a struct kmem_cache name); dev and size |
1da177e4c Linux-2.6.12-rc2 |
91 92 93 94 95 |
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. |
ad82362b2 mm: add dma_pool_... |
96 97 98 99 100 |
void *dma_pool_zalloc(struct dma_pool *pool, gfp_t mem_flags, dma_addr_t *handle) Wraps dma_pool_alloc() and also zeroes the returned memory if the allocation attempt succeeded. |
a12e2c6cd Doc: DMA-API update |
101 |
void *dma_pool_alloc(struct dma_pool *pool, gfp_t gfp_flags, |
1da177e4c Linux-2.6.12-rc2 |
102 |
dma_addr_t *dma_handle); |
77f2ea2f8 DMA-API: Clarify ... |
103 104 105 106 107 |
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 DMA-API: Capitali... |
108 |
address usable by the CPU, and the DMA address usable by the pool's |
77f2ea2f8 DMA-API: Clarify ... |
109 |
device. |
1da177e4c Linux-2.6.12-rc2 |
110 111 112 113 |
void dma_pool_free(struct dma_pool *pool, void *vaddr, dma_addr_t addr); |
1da177e4c Linux-2.6.12-rc2 |
114 |
This puts memory back into the pool. The pool is what was passed to |
f311a724a DMA-API: Capitali... |
115 |
dma_pool_alloc(); the CPU (vaddr) and DMA addresses are what |
1da177e4c Linux-2.6.12-rc2 |
116 117 118 119 |
were returned when that routine allocated the memory being freed. void dma_pool_destroy(struct dma_pool *pool); |
77f2ea2f8 DMA-API: Clarify ... |
120 |
dma_pool_destroy() frees the resources of the pool. It must be |
1da177e4c Linux-2.6.12-rc2 |
121 122 123 124 125 126 127 128 |
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 ------------------------------------ int |
4aa806b77 DMA-API: provide ... |
129 130 131 132 133 134 135 136 |
dma_set_mask_and_coherent(struct device *dev, u64 mask) 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. int |
1da177e4c Linux-2.6.12-rc2 |
137 |
dma_set_mask(struct device *dev, u64 mask) |
1da177e4c Linux-2.6.12-rc2 |
138 139 140 141 142 |
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. |
6a1961f49 dma-mapping: dma-... |
143 144 |
int dma_set_coherent_mask(struct device *dev, u64 mask) |
6a1961f49 dma-mapping: dma-... |
145 146 147 148 149 |
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. |
1da177e4c Linux-2.6.12-rc2 |
150 151 |
u64 dma_get_required_mask(struct device *dev) |
175add198 [IA64] SN specifi... |
152 153 |
This API returns the mask that the platform requires to operate efficiently. Usually this means the returned mask |
1da177e4c Linux-2.6.12-rc2 |
154 155 156 157 158 |
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 [IA64] SN specifi... |
159 160 |
wish to take advantage of it, you should issue a dma_set_mask() call to set the mask to the value returned. |
1da177e4c Linux-2.6.12-rc2 |
161 162 163 164 165 166 167 168 |
Part Id - Streaming DMA mappings -------------------------------- dma_addr_t dma_map_single(struct device *dev, void *cpu_addr, size_t size, enum dma_data_direction direction) |
1da177e4c Linux-2.6.12-rc2 |
169 170 |
Maps a piece of processor virtual memory so it can be accessed by the |
3a9ad0b4f PCI: Add pci_bus_... |
171 |
device and returns the DMA address of the memory. |
1da177e4c Linux-2.6.12-rc2 |
172 |
|
77f2ea2f8 DMA-API: Clarify ... |
173 |
The direction for both APIs may be converted freely by casting. |
1da177e4c Linux-2.6.12-rc2 |
174 175 |
However the dma_ API uses a strongly typed enumerator for its direction: |
f5a69f4c1 Documentation: re... |
176 177 178 179 |
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 |
1da177e4c Linux-2.6.12-rc2 |
180 |
|
77f2ea2f8 DMA-API: Clarify ... |
181 182 183 184 185 186 187 |
Notes: 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). |
3a9ad0b4f PCI: Add pci_bus_... |
188 |
Further, the DMA address of the memory must be within the |
77f2ea2f8 DMA-API: Clarify ... |
189 |
dma_mask of the device (the dma_mask is a bit mask of the |
3a9ad0b4f PCI: Add pci_bus_... |
190 191 |
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 |
77f2ea2f8 DMA-API: Clarify ... |
192 |
address, then the device can perform DMA to the memory). To |
1da177e4c Linux-2.6.12-rc2 |
193 |
ensure that the memory allocated by kmalloc is within the dma_mask, |
a12e2c6cd Doc: DMA-API update |
194 |
the driver may specify various platform-dependent flags to restrict |
3a9ad0b4f PCI: Add pci_bus_... |
195 196 |
the DMA address range of the allocation (e.g., on x86, GFP_DMA guarantees to be within the first 16MB of available DMA addresses, |
1da177e4c Linux-2.6.12-rc2 |
197 198 199 200 |
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 |
3a9ad0b4f PCI: Add pci_bus_... |
201 |
maps an I/O DMA address to a physical memory address). However, to be |
77f2ea2f8 DMA-API: Clarify ... |
202 203 |
portable, device driver writers may *not* assume that such an IOMMU exists. |
1da177e4c Linux-2.6.12-rc2 |
204 205 206 207 208 209 210 211 212 213 214 215 216 217 |
Warnings: 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 |
000afe892 DMA-API: fix conf... |
218 |
the device. Once this primitive is used, memory covered by this |
a12e2c6cd Doc: DMA-API update |
219 |
primitive should be treated as read-only by the device. If the device |
1da177e4c Linux-2.6.12-rc2 |
220 221 222 223 224 |
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 |
a12e2c6cd Doc: DMA-API update |
225 |
be treated as read-only by the driver. If the driver needs to write |
1da177e4c Linux-2.6.12-rc2 |
226 227 228 229 230 231 232 233 234 |
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 |
a12e2c6cd Doc: DMA-API update |
235 |
cache lines are updated with data that the device may have changed). |
1da177e4c Linux-2.6.12-rc2 |
236 237 238 239 |
void dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size, enum dma_data_direction direction) |
1da177e4c Linux-2.6.12-rc2 |
240 241 242 243 244 245 246 247 248 |
Unmaps the region previously mapped. All the parameters passed in must be identical to those passed in (and returned) by the mapping API. dma_addr_t dma_map_page(struct device *dev, struct page *page, unsigned long offset, size_t size, enum dma_data_direction direction) |
1da177e4c Linux-2.6.12-rc2 |
249 250 251 |
void dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size, enum dma_data_direction direction) |
1da177e4c Linux-2.6.12-rc2 |
252 253 254 255 256 257 258 259 |
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. int |
8d8bb39b9 dma-mapping: add ... |
260 |
dma_mapping_error(struct device *dev, dma_addr_t dma_addr) |
1da177e4c Linux-2.6.12-rc2 |
261 |
|
77f2ea2f8 DMA-API: Clarify ... |
262 |
In some circumstances dma_map_single() and dma_map_page() will fail to create |
1da177e4c Linux-2.6.12-rc2 |
263 |
a mapping. A driver can check for these errors by testing the returned |
77f2ea2f8 DMA-API: Clarify ... |
264 |
DMA address with dma_mapping_error(). A non-zero return value means the mapping |
a12e2c6cd Doc: DMA-API update |
265 |
could not be created and the driver should take appropriate action (e.g. |
1da177e4c Linux-2.6.12-rc2 |
266 |
reduce current DMA mapping usage or delay and try again later). |
21440d313 [PATCH] dma doc u... |
267 268 269 |
int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents, enum dma_data_direction direction) |
1da177e4c Linux-2.6.12-rc2 |
270 |
|
3a9ad0b4f PCI: Add pci_bus_... |
271 |
Returns: the number of DMA address segments mapped (this may be shorter |
1d678f365 DMA-API.txt: fix ... |
272 273 274 |
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 Linux-2.6.12-rc2 |
275 276 277 |
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 DMA-API: Clarify ... |
278 |
As with the other mapping interfaces, dma_map_sg() can fail. When it |
1da177e4c Linux-2.6.12-rc2 |
279 280 281 282 |
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. |
21440d313 [PATCH] dma doc u... |
283 284 285 286 |
With scatterlists, you use the resulting mapping like this: int i, count = dma_map_sg(dev, sglist, nents, direction); struct scatterlist *sg; |
79eb01457 fix an example of... |
287 |
for_each_sg(sglist, sg, count, i) { |
21440d313 [PATCH] dma doc u... |
288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 |
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. void dma_unmap_sg(struct device *dev, struct scatterlist *sg, |
7bc590b2f Documentation: DM... |
305 |
int nents, enum dma_data_direction direction) |
1da177e4c Linux-2.6.12-rc2 |
306 |
|
a12e2c6cd Doc: DMA-API update |
307 |
Unmap the previously mapped scatter/gather list. All the parameters |
1da177e4c Linux-2.6.12-rc2 |
308 309 310 311 |
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 PCI: Add pci_bus_... |
312 |
DMA address entries returned. |
1da177e4c Linux-2.6.12-rc2 |
313 |
|
9705ef7ec DMA-API.txt: add ... |
314 315 316 317 |
void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction) void |
9705ef7ec DMA-API.txt: add ... |
318 319 320 |
dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction) void |
7bc590b2f Documentation: DM... |
321 |
dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nents, |
9705ef7ec DMA-API.txt: add ... |
322 323 |
enum dma_data_direction direction) void |
7bc590b2f Documentation: DM... |
324 |
dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nents, |
9705ef7ec DMA-API.txt: add ... |
325 |
enum dma_data_direction direction) |
9705ef7ec DMA-API.txt: add ... |
326 |
|
f311a724a DMA-API: Capitali... |
327 |
Synchronise a single contiguous or scatter/gather mapping for the CPU |
9705ef7ec DMA-API.txt: add ... |
328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 |
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. Notes: 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 See also dma_map_single(). |
a75b0a2f6 dma: document dma... |
343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 |
dma_addr_t dma_map_single_attrs(struct device *dev, void *cpu_addr, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) void dma_unmap_single_attrs(struct device *dev, dma_addr_t dma_addr, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) int dma_map_sg_attrs(struct device *dev, struct scatterlist *sgl, int nents, enum dma_data_direction dir, struct dma_attrs *attrs) void dma_unmap_sg_attrs(struct device *dev, struct scatterlist *sgl, int nents, enum dma_data_direction dir, struct dma_attrs *attrs) The four functions above are just like the counterpart functions without the _attrs suffixes, except that they pass an optional struct dma_attrs*. |
77f2ea2f8 DMA-API: Clarify ... |
366 |
struct dma_attrs encapsulates a set of "DMA attributes". For the |
a75b0a2f6 dma: document dma... |
367 |
definition of struct dma_attrs see linux/dma-attrs.h. |
77f2ea2f8 DMA-API: Clarify ... |
368 |
The interpretation of DMA attributes is architecture-specific, and |
a75b0a2f6 dma: document dma... |
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 |
each attribute should be documented in Documentation/DMA-attributes.txt. If struct dma_attrs* is NULL, the semantics of each of these functions is identical to those of the corresponding function without the _attrs suffix. As a result dma_map_single_attrs() can generally replace dma_map_single(), etc. As an example of the use of the *_attrs functions, here's how you could pass an attribute DMA_ATTR_FOO when mapping memory for DMA: #include <linux/dma-attrs.h> /* DMA_ATTR_FOO should be defined in linux/dma-attrs.h and * documented in Documentation/DMA-attributes.txt */ ... DEFINE_DMA_ATTRS(attrs); dma_set_attr(DMA_ATTR_FOO, &attrs); .... n = dma_map_sg_attrs(dev, sg, nents, DMA_TO_DEVICE, &attr); .... Architectures that care about DMA_ATTR_FOO would check for its presence in their implementations of the mapping and unmapping 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, struct dma_attrs *attrs) { .... int foo = dma_get_attr(DMA_ATTR_FOO, attrs); .... if (foo) /* twizzle the frobnozzle */ .... |
1da177e4c Linux-2.6.12-rc2 |
405 406 407 |
Part II - Advanced dma_ usage ----------------------------- |
f5a69f4c1 Documentation: re... |
408 409 410 |
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 Linux-2.6.12-rc2 |
411 412 413 414 415 416 417 |
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. void * dma_alloc_noncoherent(struct device *dev, size_t size, |
a12e2c6cd Doc: DMA-API update |
418 |
dma_addr_t *dma_handle, gfp_t flag) |
1da177e4c Linux-2.6.12-rc2 |
419 420 421 422 423 424 425 426 427 428 429 |
Identical to dma_alloc_coherent() except that 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. 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 DMA-API: Clarify ... |
430 |
only use this API if you positively know your driver will be |
1da177e4c Linux-2.6.12-rc2 |
431 432 433 434 435 436 |
required to work on one of the rare (usually non-PCI) architectures that simply cannot make consistent memory. void dma_free_noncoherent(struct device *dev, size_t size, void *cpu_addr, dma_addr_t dma_handle) |
a12e2c6cd Doc: DMA-API update |
437 |
Free memory allocated by the nonconsistent API. All parameters must |
1da177e4c Linux-2.6.12-rc2 |
438 439 440 441 |
be identical to those passed in (and returned by dma_alloc_noncoherent()). int |
1da177e4c Linux-2.6.12-rc2 |
442 |
dma_get_cache_alignment(void) |
a12e2c6cd Doc: DMA-API update |
443 |
Returns the processor cache alignment. This is the absolute minimum |
1da177e4c Linux-2.6.12-rc2 |
444 445 446 447 448 449 |
alignment *and* width that you must observe when either mapping memory or doing partial flushes. Notes: 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 |
a12e2c6cd Doc: DMA-API update |
450 |
of two for easy alignment. |
1da177e4c Linux-2.6.12-rc2 |
451 452 |
void |
d3fa72e45 [PATCH] Pass stru... |
453 |
dma_cache_sync(struct device *dev, void *vaddr, size_t size, |
1da177e4c Linux-2.6.12-rc2 |
454 455 456 457 458 459 460 461 |
enum dma_data_direction direction) Do a partial sync of memory that was allocated by dma_alloc_noncoherent(), starting at virtual address vaddr and continuing on for size. Again, you *must* observe the cache line boundaries when doing this. int |
88a984ba0 DMA-API: Change d... |
462 |
dma_declare_coherent_memory(struct device *dev, phys_addr_t phys_addr, |
1da177e4c Linux-2.6.12-rc2 |
463 464 |
dma_addr_t device_addr, size_t size, int flags) |
77f2ea2f8 DMA-API: Clarify ... |
465 |
Declare region of memory to be handed out by dma_alloc_coherent() when |
1da177e4c Linux-2.6.12-rc2 |
466 |
it's asked for coherent memory for this device. |
f311a724a DMA-API: Capitali... |
467 468 |
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 Linux-2.6.12-rc2 |
469 |
|
3a9ad0b4f PCI: Add pci_bus_... |
470 |
device_addr is the DMA address the device needs to be programmed |
88a984ba0 DMA-API: Change d... |
471 |
with to actually address this memory (this will be handed out as the |
a12e2c6cd Doc: DMA-API update |
472 |
dma_addr_t in dma_alloc_coherent()). |
1da177e4c Linux-2.6.12-rc2 |
473 474 |
size is the size of the area (must be multiples of PAGE_SIZE). |
77f2ea2f8 DMA-API: Clarify ... |
475 |
flags can be ORed together and are: |
1da177e4c Linux-2.6.12-rc2 |
476 477 |
DMA_MEMORY_MAP - request that the memory returned from |
4ae0edc21 Fix typos in /Doc... |
478 |
dma_alloc_coherent() be directly writable. |
1da177e4c Linux-2.6.12-rc2 |
479 480 |
DMA_MEMORY_IO - request that the memory returned from |
77f2ea2f8 DMA-API: Clarify ... |
481 |
dma_alloc_coherent() be addressable using read()/write()/memcpy_toio() etc. |
1da177e4c Linux-2.6.12-rc2 |
482 |
|
a12e2c6cd Doc: DMA-API update |
483 |
One or both of these flags must be present. |
1da177e4c Linux-2.6.12-rc2 |
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 |
DMA_MEMORY_INCLUDES_CHILDREN - make the declared memory be allocated by dma_alloc_coherent of any child devices of this one (for memory residing on a bridge). 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. The return value will be either DMA_MEMORY_MAP or DMA_MEMORY_IO and must correspond to a passed in flag (i.e. no returning DMA_MEMORY_IO if only DMA_MEMORY_MAP were passed in) for success or zero for failure. Note, for DMA_MEMORY_IO returns, all subsequent memory returned by dma_alloc_coherent() may no longer be accessed directly, but instead must be accessed using the correct bus functions. If your driver isn't prepared to handle this contingency, it should not specify DMA_MEMORY_IO in the input flags. As a simplification for the platforms, only *one* such region of 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. void dma_release_declared_memory(struct device *dev) 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 Doc: DMA-API update |
517 |
driver's job to ensure that no parts of this memory region are |
1da177e4c Linux-2.6.12-rc2 |
518 519 520 521 522 523 524 525 |
currently in use. void * dma_mark_declared_memory_occupied(struct device *dev, dma_addr_t device_addr, size_t size) This is used to occupy specific regions of the declared space (dma_alloc_coherent() will hand out the first free region it finds). |
a12e2c6cd Doc: DMA-API update |
526 |
device_addr is the *device* address of the region requested. |
1da177e4c Linux-2.6.12-rc2 |
527 |
|
a12e2c6cd Doc: DMA-API update |
528 |
size is the size (and should be a page-sized multiple). |
1da177e4c Linux-2.6.12-rc2 |
529 530 531 532 |
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 dma-debug: Docume... |
533 534 535 |
Part III - Debug drivers use of the DMA-API ------------------------------------------- |
77f2ea2f8 DMA-API: Clarify ... |
536 |
The DMA-API as described above has some constraints. DMA addresses must be |
187f9c3f0 dma-debug: Docume... |
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 |
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 example warning message may look like this: ------------[ cut here ]------------ 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 ]--- 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: dma-api/all_errors This file contains a numeric value. If this value is not equal to zero the debugging code will print a warning for every error it finds |
19f594600 trivial: Miscella... |
599 600 |
into the kernel log. Be careful with this option, as it can easily flood your logs. |
187f9c3f0 dma-debug: Docume... |
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 |
dma-api/disabled This read-only file contains the character 'Y' if the debugging code is disabled. This can happen when it runs out of memory or if it was disabled at boot time dma-api/error_count This file is read-only and shows the total numbers of errors found. dma-api/num_errors The number in this file shows how many 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 dma-api/min_free_entries This read-only file can be read to get the 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. dma-api/num_free_entries The current number of free dma_debug_entries in the allocator. |
016ea6874 dma-debug: add do... |
626 627 628 629 630 631 |
dma-api/driver-filter You can write a name of a driver into this file 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. |
187f9c3f0 dma-debug: Docume... |
632 633 634 635 636 |
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 dma-debug: add do... |
637 638 639 640 |
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 dma-debug: Docume... |
641 642 643 644 645 |
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 dma-debug: New in... |
646 647 648 649 |
void debug_dmap_mapping_error(struct device *dev, dma_addr_t dma_addr); dma-debug interface debug_dma_mapping_error() to debug drivers that fail |
77f2ea2f8 DMA-API: Clarify ... |
650 |
to check DMA mapping errors on addresses returned by dma_map_single() and |
6c9c6d630 dma-debug: New in... |
651 652 653 654 655 |
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 DMA-API: Clarify ... |
656 |
routines to enable DMA mapping error check debugging. |
6c9c6d630 dma-debug: New in... |
657 |