Blame view
Documentation/DMA-API-HOWTO.txt
32.8 KB
266921bdb DMA-API-HOWTO.txt... |
1 2 3 |
========================= Dynamic DMA mapping Guide ========================= |
1da177e4c Linux-2.6.12-rc2 |
4 |
|
266921bdb DMA-API-HOWTO.txt... |
5 6 7 |
:Author: David S. Miller <davem@redhat.com> :Author: Richard Henderson <rth@cygnus.com> :Author: Jakub Jelinek <jakub@redhat.com> |
1da177e4c Linux-2.6.12-rc2 |
8 |
|
216bf58f4 Documentation: co... |
9 10 |
This is a guide to device driver writers on how to use the DMA API with example pseudo-code. For a concise description of the API, see |
1da177e4c Linux-2.6.12-rc2 |
11 |
DMA-API.txt. |
266921bdb DMA-API-HOWTO.txt... |
12 13 |
CPU and DMA addresses ===================== |
77f2ea2f8 DMA-API: Clarify ... |
14 15 16 17 18 19 |
There are several kinds of addresses involved in the DMA API, and it's important to understand the differences. The kernel normally uses virtual addresses. Any address returned by kmalloc(), vmalloc(), and similar interfaces is a virtual address and can |
266921bdb DMA-API-HOWTO.txt... |
20 |
be stored in a ``void *``. |
77f2ea2f8 DMA-API: Clarify ... |
21 22 23 24 25 26 27 |
The virtual memory system (TLB, page tables, etc.) translates virtual addresses to CPU physical addresses, which are stored as "phys_addr_t" or "resource_size_t". The kernel manages device resources like registers as physical addresses. These are the addresses in /proc/iomem. The physical address is not directly useful to a driver; it must use ioremap() to map the space and produce a virtual address. |
3a9ad0b4f PCI: Add pci_bus_... |
28 29 30 31 32 |
I/O devices use a third kind of address: a "bus address". If a device has registers at an MMIO address, or if it performs DMA to read or write system memory, the addresses used by the device are bus addresses. In some systems, bus addresses are identical to CPU physical addresses, but in general they are not. IOMMUs and host bridges can produce arbitrary |
77f2ea2f8 DMA-API: Clarify ... |
33 |
mappings between physical and bus addresses. |
3a9ad0b4f PCI: Add pci_bus_... |
34 35 36 37 |
From a device's point of view, DMA uses the bus address space, but it may be restricted to a subset of that space. For example, even if a system supports 64-bit addresses for main memory and PCI BARs, it may use an IOMMU so devices only need to use 32-bit DMA addresses. |
266921bdb DMA-API-HOWTO.txt... |
38 |
Here's a picture and some examples:: |
77f2ea2f8 DMA-API: Clarify ... |
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 |
CPU CPU Bus Virtual Physical Address Address Address Space Space Space +-------+ +------+ +------+ | | |MMIO | Offset | | | | Virtual |Space | applied | | C +-------+ --------> B +------+ ----------> +------+ A | | mapping | | by host | | +-----+ | | | | bridge | | +--------+ | | | | +------+ | | | | | CPU | | | | RAM | | | | Device | | | | | | | | | | | +-----+ +-------+ +------+ +------+ +--------+ | | Virtual |Buffer| Mapping | | X +-------+ --------> Y +------+ <---------- +------+ Z | | mapping | RAM | by IOMMU | | | | | | | | +-------+ +------+ During the enumeration process, the kernel learns about I/O devices and their MMIO space and the host bridges that connect them to the system. For example, if a PCI device has a BAR, the kernel reads the bus address (A) from the BAR and converts it to a CPU physical address (B). The address B is stored in a struct resource and usually exposed via /proc/iomem. When a driver claims a device, it typically uses ioremap() to map physical address B at a virtual address (C). It can then use, e.g., ioread32(C), to access the device registers at bus address A. If the device supports DMA, the driver sets up a buffer using kmalloc() or a similar interface, which returns a virtual address (X). The virtual memory system maps X to a physical address (Y) in system RAM. The driver can use virtual address X to access the buffer, but the device itself cannot because DMA doesn't go through the CPU virtual memory system. In some simple systems, the device can do DMA directly to physical address |
3a9ad0b4f PCI: Add pci_bus_... |
78 |
Y. But in many others, there is IOMMU hardware that translates DMA |
77f2ea2f8 DMA-API: Clarify ... |
79 80 81 |
addresses to physical addresses, e.g., it translates Z to Y. This is part of the reason for the DMA API: the driver can give a virtual address X to an interface like dma_map_single(), which sets up any required IOMMU |
3a9ad0b4f PCI: Add pci_bus_... |
82 |
mapping and returns the DMA address Z. The driver then tells the device to |
77f2ea2f8 DMA-API: Clarify ... |
83 84 |
do DMA to Z, and the IOMMU maps it to the buffer at address Y in system RAM. |
1da177e4c Linux-2.6.12-rc2 |
85 86 87 88 89 90 91 |
So that Linux can use the dynamic DMA mapping, it needs some help from the drivers, namely it has to take into account that DMA addresses should be mapped only for the time they are actually used and unmapped after the DMA transfer. The following API will work of course even on platforms where no such |
216bf58f4 Documentation: co... |
92 93 94 |
hardware exists. Note that the DMA API works with any bus independent of the underlying |
77f2ea2f8 DMA-API: Clarify ... |
95 96 97 |
microprocessor architecture. You should use the DMA API rather than the bus-specific DMA API, i.e., use the dma_map_*() interfaces rather than the pci_map_*() interfaces. |
1da177e4c Linux-2.6.12-rc2 |
98 |
|
266921bdb DMA-API-HOWTO.txt... |
99 |
First of all, you should make sure:: |
1da177e4c Linux-2.6.12-rc2 |
100 |
|
266921bdb DMA-API-HOWTO.txt... |
101 |
#include <linux/dma-mapping.h> |
1da177e4c Linux-2.6.12-rc2 |
102 |
|
77f2ea2f8 DMA-API: Clarify ... |
103 |
is in your driver, which provides the definition of dma_addr_t. This type |
3a9ad0b4f PCI: Add pci_bus_... |
104 |
can hold any valid DMA address for the platform and should be used |
77f2ea2f8 DMA-API: Clarify ... |
105 |
everywhere you hold a DMA address returned from the DMA mapping functions. |
1da177e4c Linux-2.6.12-rc2 |
106 |
|
266921bdb DMA-API-HOWTO.txt... |
107 108 |
What memory is DMA'able? ======================== |
1da177e4c Linux-2.6.12-rc2 |
109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 |
The first piece of information you must know is what kernel memory can be used with the DMA mapping facilities. There has been an unwritten set of rules regarding this, and this text is an attempt to finally write them down. If you acquired your memory via the page allocator (i.e. __get_free_page*()) or the generic memory allocators (i.e. kmalloc() or kmem_cache_alloc()) then you may DMA to/from that memory using the addresses returned from those routines. This means specifically that you may _not_ use the memory/addresses returned from vmalloc() for DMA. It is possible to DMA to the _underlying_ memory mapped into a vmalloc() area, but this requires walking page tables to get the physical addresses, and then translating each of those pages back to a kernel address using something like __va(). [ EDIT: Update this when we integrate Gerd Knorr's generic code which does this. ] |
21440d313 [PATCH] dma doc u... |
127 128 129 130 131 132 133 134 135 |
This rule also means that you may use neither kernel image addresses (items in data/text/bss segments), nor module image addresses, nor stack addresses for DMA. These could all be mapped somewhere entirely different than the rest of physical memory. Even if those classes of memory could physically work with DMA, you'd need to ensure the I/O buffers were cacheline-aligned. Without that, you'd see cacheline sharing problems (data corruption) on CPUs with DMA-incoherent caches. (The CPU could write to one word, DMA would write to a different one in the same cache line, and one of them could be overwritten.) |
1da177e4c Linux-2.6.12-rc2 |
136 137 138 139 140 141 142 |
Also, this means that you cannot take the return of a kmap() call and DMA to/from that. This is similar to vmalloc(). What about block I/O and networking buffers? The block I/O and networking subsystems make sure that the buffers they use are valid for you to DMA from/to. |
9eb9e96e9 Documentation/DMA... |
143 |
DMA addressing capabilities |
9fda5130d docs: DMA-API-HOW... |
144 |
=========================== |
1da177e4c Linux-2.6.12-rc2 |
145 |
|
9eb9e96e9 Documentation/DMA... |
146 147 148 |
By default, the kernel assumes that your device can address 32-bits of DMA addressing. For a 64-bit capable device, this needs to be increased, and for a device with limitations, it needs to be decreased. |
1da177e4c Linux-2.6.12-rc2 |
149 |
|
9eb9e96e9 Documentation/DMA... |
150 151 152 153 |
Special note about PCI: PCI-X specification requires PCI-X devices to support 64-bit addressing (DAC) for all transactions. And at least one platform (SGI SN2) requires 64-bit consistent allocations to operate correctly when the IO bus is in PCI-X mode. |
216bf58f4 Documentation: co... |
154 |
|
9eb9e96e9 Documentation/DMA... |
155 156 |
For correct operation, you must set the DMA mask to inform the kernel about your devices DMA addressing capabilities. |
216bf58f4 Documentation: co... |
157 |
|
9eb9e96e9 Documentation/DMA... |
158 |
This is performed via a call to dma_set_mask_and_coherent():: |
1da177e4c Linux-2.6.12-rc2 |
159 |
|
4aa806b77 DMA-API: provide ... |
160 |
int dma_set_mask_and_coherent(struct device *dev, u64 mask); |
1da177e4c Linux-2.6.12-rc2 |
161 |
|
9eb9e96e9 Documentation/DMA... |
162 163 164 |
which will set the mask for both streaming and coherent APIs together. If you have some special requirements, then the following two separate calls can be used instead: |
1da177e4c Linux-2.6.12-rc2 |
165 |
|
9eb9e96e9 Documentation/DMA... |
166 |
The setup for streaming mappings is performed via a call to |
266921bdb DMA-API-HOWTO.txt... |
167 |
dma_set_mask():: |
4aa806b77 DMA-API: provide ... |
168 169 |
int dma_set_mask(struct device *dev, u64 mask); |
9eb9e96e9 Documentation/DMA... |
170 |
The setup for consistent allocations is performed via a call |
266921bdb DMA-API-HOWTO.txt... |
171 |
to dma_set_coherent_mask():: |
4aa806b77 DMA-API: provide ... |
172 173 |
int dma_set_coherent_mask(struct device *dev, u64 mask); |
1da177e4c Linux-2.6.12-rc2 |
174 |
|
9eb9e96e9 Documentation/DMA... |
175 176 177 178 179 |
Here, dev is a pointer to the device struct of your device, and mask is a bit mask describing which bits of an address your device supports. Often the device struct of your device is embedded in the bus-specific device struct of your device. For example, &pdev->dev is a pointer to the device struct of a PCI device (pdev is a pointer to the PCI device struct of your device). |
1da177e4c Linux-2.6.12-rc2 |
180 |
|
9eb9e96e9 Documentation/DMA... |
181 182 183 184 185 186 187 |
These calls usually return zero to indicated your device can perform DMA properly on the machine given the address mask you provided, but they might return an error if the mask is too small to be supportable on the given system. If it returns non-zero, your device cannot perform DMA properly on this platform, and attempting to do so will result in undefined behavior. You must not use DMA on this device unless the dma_set_mask family of functions has returned success. |
1da177e4c Linux-2.6.12-rc2 |
188 |
|
9eb9e96e9 Documentation/DMA... |
189 |
This means that in the failure case, you have two options: |
1da177e4c Linux-2.6.12-rc2 |
190 |
|
9eb9e96e9 Documentation/DMA... |
191 192 |
1) Use some non-DMA mode for data transfer, if possible. 2) Ignore this device and do not initialize it. |
1da177e4c Linux-2.6.12-rc2 |
193 |
|
9eb9e96e9 Documentation/DMA... |
194 195 196 197 |
It is recommended that your driver print a kernel KERN_WARNING message when setting the DMA mask fails. In this manner, if a user of your driver reports that performance is bad or that the device is not even detected, you can ask them for the kernel messages to find out exactly why. |
1da177e4c Linux-2.6.12-rc2 |
198 |
|
9eb9e96e9 Documentation/DMA... |
199 |
The standard 64-bit addressing device would do something like this:: |
1da177e4c Linux-2.6.12-rc2 |
200 |
|
9eb9e96e9 Documentation/DMA... |
201 |
if (dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64))) { |
77f2ea2f8 DMA-API: Clarify ... |
202 203 |
dev_warn(dev, "mydev: No suitable DMA available "); |
1da177e4c Linux-2.6.12-rc2 |
204 205 |
goto ignore_this_device; } |
9eb9e96e9 Documentation/DMA... |
206 207 |
If the device only supports 32-bit addressing for descriptors in the coherent allocations, but supports full 64-bits for streaming mappings |
a36d05386 docs: DMA-API-HOW... |
208 |
it would look like this:: |
1da177e4c Linux-2.6.12-rc2 |
209 |
|
9eb9e96e9 Documentation/DMA... |
210 |
if (dma_set_mask(dev, DMA_BIT_MASK(64))) { |
77f2ea2f8 DMA-API: Clarify ... |
211 212 |
dev_warn(dev, "mydev: No suitable DMA available "); |
1da177e4c Linux-2.6.12-rc2 |
213 214 |
goto ignore_this_device; } |
34c815fbd DMA-API: Fix dupl... |
215 216 217 218 |
The coherent mask will always be able to set the same or a smaller mask as the streaming mask. However for the rare case that a device driver only uses consistent allocations, one would have to check the return value from dma_set_coherent_mask(). |
1da177e4c Linux-2.6.12-rc2 |
219 |
|
1da177e4c Linux-2.6.12-rc2 |
220 |
Finally, if your device can only drive the low 24-bits of |
266921bdb DMA-API-HOWTO.txt... |
221 |
address you might do something like:: |
1da177e4c Linux-2.6.12-rc2 |
222 |
|
216bf58f4 Documentation: co... |
223 |
if (dma_set_mask(dev, DMA_BIT_MASK(24))) { |
77f2ea2f8 DMA-API: Clarify ... |
224 225 |
dev_warn(dev, "mydev: 24-bit DMA addressing not available "); |
1da177e4c Linux-2.6.12-rc2 |
226 227 |
goto ignore_this_device; } |
4aa806b77 DMA-API: provide ... |
228 229 230 |
When dma_set_mask() or dma_set_mask_and_coherent() is successful, and returns zero, the kernel saves away this mask you have provided. The kernel will use this information later when you make DMA mappings. |
1da177e4c Linux-2.6.12-rc2 |
231 232 233 234 235 236 237 |
There is a case which we are aware of at this time, which is worth mentioning in this documentation. If your device supports multiple functions (for example a sound card provides playback and record functions) and the various different functions have _different_ DMA addressing limitations, you may wish to probe each mask and only provide the functionality which the machine can handle. It |
216bf58f4 Documentation: co... |
238 |
is important that the last call to dma_set_mask() be for the |
1da177e4c Linux-2.6.12-rc2 |
239 |
most specific mask. |
266921bdb DMA-API-HOWTO.txt... |
240 |
Here is pseudo-code showing how this might be done:: |
1da177e4c Linux-2.6.12-rc2 |
241 |
|
2c5510d4e dma-mapping: upda... |
242 |
#define PLAYBACK_ADDRESS_BITS DMA_BIT_MASK(32) |
038f7d002 use DMA_BIT_MASK ... |
243 |
#define RECORD_ADDRESS_BITS DMA_BIT_MASK(24) |
1da177e4c Linux-2.6.12-rc2 |
244 245 |
struct my_sound_card *card; |
216bf58f4 Documentation: co... |
246 |
struct device *dev; |
1da177e4c Linux-2.6.12-rc2 |
247 248 |
... |
216bf58f4 Documentation: co... |
249 |
if (!dma_set_mask(dev, PLAYBACK_ADDRESS_BITS)) { |
1da177e4c Linux-2.6.12-rc2 |
250 251 252 |
card->playback_enabled = 1; } else { card->playback_enabled = 0; |
77f2ea2f8 DMA-API: Clarify ... |
253 254 |
dev_warn(dev, "%s: Playback disabled due to DMA limitations ", |
1da177e4c Linux-2.6.12-rc2 |
255 256 |
card->name); } |
216bf58f4 Documentation: co... |
257 |
if (!dma_set_mask(dev, RECORD_ADDRESS_BITS)) { |
1da177e4c Linux-2.6.12-rc2 |
258 259 260 |
card->record_enabled = 1; } else { card->record_enabled = 0; |
77f2ea2f8 DMA-API: Clarify ... |
261 262 |
dev_warn(dev, "%s: Record disabled due to DMA limitations ", |
1da177e4c Linux-2.6.12-rc2 |
263 264 265 266 267 268 |
card->name); } A sound card was used as an example here because this genre of PCI devices seems to be littered with ISA chips given a PCI front end, and thus retaining the 16MB DMA addressing limitations of ISA. |
266921bdb DMA-API-HOWTO.txt... |
269 270 |
Types of DMA mappings ===================== |
1da177e4c Linux-2.6.12-rc2 |
271 272 273 274 275 276 277 278 279 280 281 282 |
There are two types of DMA mappings: - Consistent DMA mappings which are usually mapped at driver initialization, unmapped at the end and for which the hardware should guarantee that the device and the CPU can access the data in parallel and will see updates made by each other without any explicit software flushing. Think of "consistent" as "synchronous" or "coherent". The current default is to return consistent memory in the low 32 |
3a9ad0b4f PCI: Add pci_bus_... |
283 |
bits of the DMA space. However, for future compatibility you should |
216bf58f4 Documentation: co... |
284 |
set the consistent mask even if this default is fine for your |
1da177e4c Linux-2.6.12-rc2 |
285 286 287 288 289 290 291 292 293 294 295 296 |
driver. Good examples of what to use consistent mappings for are: - Network card DMA ring descriptors. - SCSI adapter mailbox command data structures. - Device firmware microcode executed out of main memory. The invariant these examples all require is that any CPU store to memory is immediately visible to the device, and vice versa. Consistent mappings guarantee this. |
266921bdb DMA-API-HOWTO.txt... |
297 298 299 300 |
.. important:: Consistent DMA memory does not preclude the usage of proper memory barriers. The CPU may reorder stores to |
1da177e4c Linux-2.6.12-rc2 |
301 302 303 |
consistent memory just as it may normal memory. Example: if it is important for the device to see the first word of a descriptor updated before the second, you must do |
266921bdb DMA-API-HOWTO.txt... |
304 |
something like:: |
1da177e4c Linux-2.6.12-rc2 |
305 306 307 308 309 310 |
desc->word0 = address; wmb(); desc->word1 = DESC_VALID; in order to get correct behavior on all platforms. |
21440d313 [PATCH] dma doc u... |
311 312 313 314 |
Also, on some platforms your driver may need to flush CPU write buffers in much the same way as it needs to flush write buffers found in PCI bridges (such as by reading a register's value after writing it). |
216bf58f4 Documentation: co... |
315 316 317 |
- Streaming DMA mappings which are usually mapped for one DMA transfer, unmapped right after it (unless you use dma_sync_* below) and for which hardware can optimize for sequential accesses. |
1da177e4c Linux-2.6.12-rc2 |
318 |
|
11e285d8f DMA-API: Spelling... |
319 |
Think of "streaming" as "asynchronous" or "outside the coherency |
1da177e4c Linux-2.6.12-rc2 |
320 321 322 323 324 325 326 327 328 329 330 |
domain". Good examples of what to use streaming mappings for are: - Networking buffers transmitted/received by a device. - Filesystem buffers written/read by a SCSI device. The interfaces for using this type of mapping were designed in such a way that an implementation can make whatever performance optimizations the hardware allows. To this end, when using such mappings you must be explicit about what you want to happen. |
216bf58f4 Documentation: co... |
331 332 |
Neither type of DMA mapping has alignment restrictions that come from the underlying bus, although some devices may have such restrictions. |
21440d313 [PATCH] dma doc u... |
333 334 |
Also, systems with caches that aren't DMA-coherent will work better when the underlying buffers don't share cache lines with other data. |
1da177e4c Linux-2.6.12-rc2 |
335 |
|
266921bdb DMA-API-HOWTO.txt... |
336 337 |
Using Consistent DMA mappings ============================= |
1da177e4c Linux-2.6.12-rc2 |
338 339 |
To allocate and map large (PAGE_SIZE or so) consistent DMA regions, |
266921bdb DMA-API-HOWTO.txt... |
340 |
you should do:: |
1da177e4c Linux-2.6.12-rc2 |
341 342 |
dma_addr_t dma_handle; |
216bf58f4 Documentation: co... |
343 |
cpu_addr = dma_alloc_coherent(dev, size, &dma_handle, gfp); |
1da177e4c Linux-2.6.12-rc2 |
344 |
|
266921bdb DMA-API-HOWTO.txt... |
345 |
where device is a ``struct device *``. This may be called in interrupt |
216bf58f4 Documentation: co... |
346 |
context with the GFP_ATOMIC flag. |
1da177e4c Linux-2.6.12-rc2 |
347 348 349 350 |
Size is the length of the region you want to allocate, in bytes. This routine will allocate RAM for that region, so it acts similarly to |
77f2ea2f8 DMA-API: Clarify ... |
351 |
__get_free_pages() (but takes size instead of a page order). If your |
1da177e4c Linux-2.6.12-rc2 |
352 |
driver needs regions sized smaller than a page, you may prefer using |
216bf58f4 Documentation: co... |
353 |
the dma_pool interface, described below. |
d7e02a931 dma-mapping: remo... |
354 355 356 357 358 359 |
The consistent DMA mapping interfaces, will by default return a DMA address which is 32-bit addressable. Even if the device indicates (via the DMA mask) that it may address the upper 32-bits, consistent allocation will only return > 32-bit addresses for DMA if the consistent DMA mask has been explicitly changed via dma_set_coherent_mask(). This is true of the dma_pool interface as well. |
216bf58f4 Documentation: co... |
360 |
|
77f2ea2f8 DMA-API: Clarify ... |
361 |
dma_alloc_coherent() returns two values: the virtual address which you |
1da177e4c Linux-2.6.12-rc2 |
362 363 |
can use to access it from the CPU and dma_handle which you pass to the card. |
3a9ad0b4f PCI: Add pci_bus_... |
364 |
The CPU virtual address and the DMA address are both |
1da177e4c Linux-2.6.12-rc2 |
365 366 367 368 369 |
guaranteed to be aligned to the smallest PAGE_SIZE order which is greater than or equal to the requested size. This invariant exists (for example) to guarantee that if you allocate a chunk which is smaller than or equal to 64 kilobytes, the extent of the buffer you receive will not cross a 64K boundary. |
266921bdb DMA-API-HOWTO.txt... |
370 |
To unmap and free such a DMA region, you call:: |
1da177e4c Linux-2.6.12-rc2 |
371 |
|
216bf58f4 Documentation: co... |
372 |
dma_free_coherent(dev, size, cpu_addr, dma_handle); |
1da177e4c Linux-2.6.12-rc2 |
373 |
|
216bf58f4 Documentation: co... |
374 |
where dev, size are the same as in the above call and cpu_addr and |
77f2ea2f8 DMA-API: Clarify ... |
375 |
dma_handle are the values dma_alloc_coherent() returned to you. |
1da177e4c Linux-2.6.12-rc2 |
376 377 378 |
This function may not be called in interrupt context. If your driver needs lots of smaller memory regions, you can write |
77f2ea2f8 DMA-API: Clarify ... |
379 |
custom code to subdivide pages returned by dma_alloc_coherent(), |
216bf58f4 Documentation: co... |
380 |
or you can use the dma_pool API to do that. A dma_pool is like |
77f2ea2f8 DMA-API: Clarify ... |
381 |
a kmem_cache, but it uses dma_alloc_coherent(), not __get_free_pages(). |
1da177e4c Linux-2.6.12-rc2 |
382 383 |
Also, it understands common hardware constraints for alignment, like queue heads needing to be aligned on N byte boundaries. |
266921bdb DMA-API-HOWTO.txt... |
384 |
Create a dma_pool like this:: |
1da177e4c Linux-2.6.12-rc2 |
385 |
|
216bf58f4 Documentation: co... |
386 |
struct dma_pool *pool; |
1da177e4c Linux-2.6.12-rc2 |
387 |
|
2af9da863 DMA-API: Update d... |
388 |
pool = dma_pool_create(name, dev, size, align, boundary); |
1da177e4c Linux-2.6.12-rc2 |
389 |
|
216bf58f4 Documentation: co... |
390 |
The "name" is for diagnostics (like a kmem_cache name); dev and size |
1da177e4c Linux-2.6.12-rc2 |
391 392 393 |
are as above. 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, |
2af9da863 DMA-API: Update d... |
394 |
pass 0 for boundary; passing 4096 says memory allocated from this pool |
1da177e4c Linux-2.6.12-rc2 |
395 |
must not cross 4KByte boundaries (but at that time it may be better to |
77f2ea2f8 DMA-API: Clarify ... |
396 |
use dma_alloc_coherent() directly instead). |
1da177e4c Linux-2.6.12-rc2 |
397 |
|
266921bdb DMA-API-HOWTO.txt... |
398 |
Allocate memory from a DMA pool like this:: |
1da177e4c Linux-2.6.12-rc2 |
399 |
|
216bf58f4 Documentation: co... |
400 |
cpu_addr = dma_pool_alloc(pool, flags, &dma_handle); |
1da177e4c Linux-2.6.12-rc2 |
401 |
|
2af9da863 DMA-API: Update d... |
402 403 |
flags are GFP_KERNEL if blocking is permitted (not in_interrupt nor holding SMP locks), GFP_ATOMIC otherwise. Like dma_alloc_coherent(), |
1da177e4c Linux-2.6.12-rc2 |
404 |
this returns two values, cpu_addr and dma_handle. |
266921bdb DMA-API-HOWTO.txt... |
405 |
Free memory that was allocated from a dma_pool like this:: |
1da177e4c Linux-2.6.12-rc2 |
406 |
|
216bf58f4 Documentation: co... |
407 |
dma_pool_free(pool, cpu_addr, dma_handle); |
1da177e4c Linux-2.6.12-rc2 |
408 |
|
77f2ea2f8 DMA-API: Clarify ... |
409 410 |
where pool is what you passed to dma_pool_alloc(), and cpu_addr and dma_handle are the values dma_pool_alloc() returned. This function |
1da177e4c Linux-2.6.12-rc2 |
411 |
may be called in interrupt context. |
266921bdb DMA-API-HOWTO.txt... |
412 |
Destroy a dma_pool by calling:: |
1da177e4c Linux-2.6.12-rc2 |
413 |
|
216bf58f4 Documentation: co... |
414 |
dma_pool_destroy(pool); |
1da177e4c Linux-2.6.12-rc2 |
415 |
|
77f2ea2f8 DMA-API: Clarify ... |
416 |
Make sure you've called dma_pool_free() for all memory allocated |
1da177e4c Linux-2.6.12-rc2 |
417 418 |
from a pool before you destroy the pool. This function may not be called in interrupt context. |
266921bdb DMA-API-HOWTO.txt... |
419 420 |
DMA Direction ============= |
1da177e4c Linux-2.6.12-rc2 |
421 422 423 |
The interfaces described in subsequent portions of this document take a DMA direction argument, which is an integer and takes on |
266921bdb DMA-API-HOWTO.txt... |
424 |
one of the following values:: |
1da177e4c Linux-2.6.12-rc2 |
425 |
|
216bf58f4 Documentation: co... |
426 427 428 429 |
DMA_BIDIRECTIONAL DMA_TO_DEVICE DMA_FROM_DEVICE DMA_NONE |
1da177e4c Linux-2.6.12-rc2 |
430 |
|
77f2ea2f8 DMA-API: Clarify ... |
431 |
You should provide the exact DMA direction if you know it. |
1da177e4c Linux-2.6.12-rc2 |
432 |
|
216bf58f4 Documentation: co... |
433 434 |
DMA_TO_DEVICE means "from main memory to the device" DMA_FROM_DEVICE means "from the device to main memory" |
1da177e4c Linux-2.6.12-rc2 |
435 436 437 438 439 440 441 |
It is the direction in which the data moves during the DMA transfer. You are _strongly_ encouraged to specify this as precisely as you possibly can. If you absolutely cannot know the direction of the DMA transfer, |
216bf58f4 Documentation: co... |
442 |
specify DMA_BIDIRECTIONAL. It means that the DMA can go in |
1da177e4c Linux-2.6.12-rc2 |
443 444 445 |
either direction. The platform guarantees that you may legally specify this, and that it will work, but this may be at the cost of performance for example. |
216bf58f4 Documentation: co... |
446 |
The value DMA_NONE is to be used for debugging. One can |
1da177e4c Linux-2.6.12-rc2 |
447 448 449 450 451 452 453 454 455 |
hold this in a data structure before you come to know the precise direction, and this will help catch cases where your direction tracking logic has failed to set things up properly. Another advantage of specifying this value precisely (outside of potential platform-specific optimizations of such) is for debugging. Some platforms actually have a write permission boolean which DMA mappings can be marked with, much like page protections in the user program address space. Such platforms can and do report errors in the |
216bf58f4 Documentation: co... |
456 |
kernel logs when the DMA controller hardware detects violation of the |
1da177e4c Linux-2.6.12-rc2 |
457 458 459 460 |
permission setting. Only streaming mappings specify a direction, consistent mappings implicitly have a direction attribute setting of |
216bf58f4 Documentation: co... |
461 |
DMA_BIDIRECTIONAL. |
1da177e4c Linux-2.6.12-rc2 |
462 |
|
be7db055d [PATCH] remove ol... |
463 464 465 |
The SCSI subsystem tells you the direction to use in the 'sc_data_direction' member of the SCSI command your driver is working on. |
1da177e4c Linux-2.6.12-rc2 |
466 467 |
For Networking drivers, it's a rather simple affair. For transmit |
216bf58f4 Documentation: co... |
468 |
packets, map/unmap them with the DMA_TO_DEVICE direction |
1da177e4c Linux-2.6.12-rc2 |
469 |
specifier. For receive packets, just the opposite, map/unmap them |
216bf58f4 Documentation: co... |
470 |
with the DMA_FROM_DEVICE direction specifier. |
1da177e4c Linux-2.6.12-rc2 |
471 |
|
266921bdb DMA-API-HOWTO.txt... |
472 473 |
Using Streaming DMA mappings ============================ |
1da177e4c Linux-2.6.12-rc2 |
474 475 476 477 478 |
The streaming DMA mapping routines can be called from interrupt context. There are two versions of each map/unmap, one which will map/unmap a single memory region, and one which will map/unmap a scatterlist. |
266921bdb DMA-API-HOWTO.txt... |
479 |
To map a single region, you do:: |
1da177e4c Linux-2.6.12-rc2 |
480 |
|
216bf58f4 Documentation: co... |
481 |
struct device *dev = &my_dev->dev; |
1da177e4c Linux-2.6.12-rc2 |
482 483 484 |
dma_addr_t dma_handle; void *addr = buffer->ptr; size_t size = buffer->len; |
216bf58f4 Documentation: co... |
485 |
dma_handle = dma_map_single(dev, addr, size, direction); |
b2dd83b37 Documentation: co... |
486 |
if (dma_mapping_error(dev, dma_handle)) { |
8d7f62e6a Documentation DMA... |
487 488 489 490 491 492 493 |
/* * reduce current DMA mapping usage, * delay and try again later or * reset driver. */ goto map_error_handling; } |
1da177e4c Linux-2.6.12-rc2 |
494 |
|
266921bdb DMA-API-HOWTO.txt... |
495 |
and to unmap it:: |
1da177e4c Linux-2.6.12-rc2 |
496 |
|
216bf58f4 Documentation: co... |
497 |
dma_unmap_single(dev, dma_handle, size, direction); |
1da177e4c Linux-2.6.12-rc2 |
498 |
|
8d7f62e6a Documentation DMA... |
499 |
You should call dma_mapping_error() as dma_map_single() could fail and return |
f51f288e2 dma-mapping: remo... |
500 501 502 503 504 |
error. Doing so will ensure that the mapping code will work correctly on all DMA implementations without any dependency on the specifics of the underlying implementation. Using the returned address without checking for errors could result in failures ranging from panics to silent data corruption. The same applies to dma_map_page() as well. |
8d7f62e6a Documentation DMA... |
505 |
|
77f2ea2f8 DMA-API: Clarify ... |
506 |
You should call dma_unmap_single() when the DMA activity is finished, e.g., |
1da177e4c Linux-2.6.12-rc2 |
507 |
from the interrupt which told you that the DMA transfer is done. |
f311a724a DMA-API: Capitali... |
508 |
Using CPU pointers like this for single mappings has a disadvantage: |
1da177e4c Linux-2.6.12-rc2 |
509 |
you cannot reference HIGHMEM memory in this way. Thus, there is a |
77f2ea2f8 DMA-API: Clarify ... |
510 |
map/unmap interface pair akin to dma_{map,unmap}_single(). These |
f311a724a DMA-API: Capitali... |
511 |
interfaces deal with page/offset pairs instead of CPU pointers. |
266921bdb DMA-API-HOWTO.txt... |
512 |
Specifically:: |
1da177e4c Linux-2.6.12-rc2 |
513 |
|
216bf58f4 Documentation: co... |
514 |
struct device *dev = &my_dev->dev; |
1da177e4c Linux-2.6.12-rc2 |
515 516 517 518 |
dma_addr_t dma_handle; struct page *page = buffer->page; unsigned long offset = buffer->offset; size_t size = buffer->len; |
216bf58f4 Documentation: co... |
519 |
dma_handle = dma_map_page(dev, page, offset, size, direction); |
b2dd83b37 Documentation: co... |
520 |
if (dma_mapping_error(dev, dma_handle)) { |
8d7f62e6a Documentation DMA... |
521 522 523 524 525 526 527 |
/* * reduce current DMA mapping usage, * delay and try again later or * reset driver. */ goto map_error_handling; } |
1da177e4c Linux-2.6.12-rc2 |
528 529 |
... |
216bf58f4 Documentation: co... |
530 |
dma_unmap_page(dev, dma_handle, size, direction); |
1da177e4c Linux-2.6.12-rc2 |
531 532 |
Here, "offset" means byte offset within the given page. |
8d7f62e6a Documentation DMA... |
533 534 |
You should call dma_mapping_error() as dma_map_page() could fail and return error as outlined under the dma_map_single() discussion. |
77f2ea2f8 DMA-API: Clarify ... |
535 |
You should call dma_unmap_page() when the DMA activity is finished, e.g., |
8d7f62e6a Documentation DMA... |
536 |
from the interrupt which told you that the DMA transfer is done. |
266921bdb DMA-API-HOWTO.txt... |
537 |
With scatterlists, you map a region gathered from several regions by:: |
1da177e4c Linux-2.6.12-rc2 |
538 |
|
216bf58f4 Documentation: co... |
539 |
int i, count = dma_map_sg(dev, sglist, nents, direction); |
1da177e4c Linux-2.6.12-rc2 |
540 |
struct scatterlist *sg; |
4c2f6d4c2 use sg helper fun... |
541 |
for_each_sg(sglist, sg, count, i) { |
1da177e4c Linux-2.6.12-rc2 |
542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 |
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. if DMA mapping is done with PAGE_SIZE granularity, any consecutive sglist entries can be merged into one provided the first one ends and the second one starts on a page boundary - in fact this is a huge advantage for cards which either cannot do scatter-gather or have very limited number of scatter-gather entries) 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. |
266921bdb DMA-API-HOWTO.txt... |
559 |
To unmap a scatterlist, just call:: |
1da177e4c Linux-2.6.12-rc2 |
560 |
|
216bf58f4 Documentation: co... |
561 |
dma_unmap_sg(dev, sglist, nents, direction); |
1da177e4c Linux-2.6.12-rc2 |
562 563 |
Again, make sure DMA activity has already finished. |
266921bdb DMA-API-HOWTO.txt... |
564 565 566 567 568 569 |
.. note:: The 'nents' argument to the dma_unmap_sg call must be the _same_ one you passed into the dma_map_sg call, it should _NOT_ be the 'count' value _returned_ from the dma_map_sg call. |
1da177e4c Linux-2.6.12-rc2 |
570 |
|
77f2ea2f8 DMA-API: Clarify ... |
571 |
Every dma_map_{single,sg}() call should have its dma_unmap_{single,sg}() |
3a9ad0b4f PCI: Add pci_bus_... |
572 573 |
counterpart, because the DMA address space is a shared resource and you could render the machine unusable by consuming all DMA addresses. |
1da177e4c Linux-2.6.12-rc2 |
574 575 576 |
If you need to use the same streaming DMA region multiple times and touch the data in between the DMA transfers, the buffer needs to be synced |
f311a724a DMA-API: Capitali... |
577 |
properly in order for the CPU and device to see the most up-to-date and |
1da177e4c Linux-2.6.12-rc2 |
578 |
correct copy of the DMA buffer. |
77f2ea2f8 DMA-API: Clarify ... |
579 |
So, firstly, just map it with dma_map_{single,sg}(), and after each DMA |
266921bdb DMA-API-HOWTO.txt... |
580 |
transfer call either:: |
1da177e4c Linux-2.6.12-rc2 |
581 |
|
216bf58f4 Documentation: co... |
582 |
dma_sync_single_for_cpu(dev, dma_handle, size, direction); |
1da177e4c Linux-2.6.12-rc2 |
583 |
|
266921bdb DMA-API-HOWTO.txt... |
584 |
or:: |
1da177e4c Linux-2.6.12-rc2 |
585 |
|
216bf58f4 Documentation: co... |
586 |
dma_sync_sg_for_cpu(dev, sglist, nents, direction); |
1da177e4c Linux-2.6.12-rc2 |
587 588 589 590 |
as appropriate. Then, if you wish to let the device get at the DMA area again, |
f311a724a DMA-API: Capitali... |
591 |
finish accessing the data with the CPU, and then before actually |
266921bdb DMA-API-HOWTO.txt... |
592 |
giving the buffer to the hardware call either:: |
1da177e4c Linux-2.6.12-rc2 |
593 |
|
216bf58f4 Documentation: co... |
594 |
dma_sync_single_for_device(dev, dma_handle, size, direction); |
1da177e4c Linux-2.6.12-rc2 |
595 |
|
266921bdb DMA-API-HOWTO.txt... |
596 |
or:: |
1da177e4c Linux-2.6.12-rc2 |
597 |
|
216bf58f4 Documentation: co... |
598 |
dma_sync_sg_for_device(dev, sglist, nents, direction); |
1da177e4c Linux-2.6.12-rc2 |
599 600 |
as appropriate. |
266921bdb DMA-API-HOWTO.txt... |
601 602 603 |
.. note:: The 'nents' argument to dma_sync_sg_for_cpu() and |
7bc590b2f Documentation: DM... |
604 605 606 |
dma_sync_sg_for_device() must be the same passed to dma_map_sg(). It is _NOT_ the count returned by dma_map_sg(). |
1da177e4c Linux-2.6.12-rc2 |
607 |
After the last DMA transfer call one of the DMA unmap routines |
77f2ea2f8 DMA-API: Clarify ... |
608 609 610 |
dma_unmap_{single,sg}(). If you don't touch the data from the first dma_map_*() call till dma_unmap_*(), then you don't have to call the dma_sync_*() routines at all. |
1da177e4c Linux-2.6.12-rc2 |
611 612 |
Here is pseudo code which shows a situation in which you would need |
266921bdb DMA-API-HOWTO.txt... |
613 |
to use the dma_sync_*() interfaces:: |
1da177e4c Linux-2.6.12-rc2 |
614 615 616 617 |
my_card_setup_receive_buffer(struct my_card *cp, char *buffer, int len) { dma_addr_t mapping; |
216bf58f4 Documentation: co... |
618 |
mapping = dma_map_single(cp->dev, buffer, len, DMA_FROM_DEVICE); |
be6c30956 Documentation: DM... |
619 |
if (dma_mapping_error(cp->dev, mapping)) { |
8d7f62e6a Documentation DMA... |
620 621 622 623 624 625 626 |
/* * reduce current DMA mapping usage, * delay and try again later or * reset driver. */ goto map_error_handling; } |
1da177e4c Linux-2.6.12-rc2 |
627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 |
cp->rx_buf = buffer; cp->rx_len = len; cp->rx_dma = mapping; give_rx_buf_to_card(cp); } ... my_card_interrupt_handler(int irq, void *devid, struct pt_regs *regs) { struct my_card *cp = devid; ... if (read_card_status(cp) == RX_BUF_TRANSFERRED) { struct my_card_header *hp; /* Examine the header to see if we wish * to accept the data. But synchronize * the DMA transfer with the CPU first * so that we see updated contents. */ |
216bf58f4 Documentation: co... |
650 651 652 |
dma_sync_single_for_cpu(&cp->dev, cp->rx_dma, cp->rx_len, DMA_FROM_DEVICE); |
1da177e4c Linux-2.6.12-rc2 |
653 654 655 656 |
/* Now it is safe to examine the buffer. */ hp = (struct my_card_header *) cp->rx_buf; if (header_is_ok(hp)) { |
216bf58f4 Documentation: co... |
657 658 |
dma_unmap_single(&cp->dev, cp->rx_dma, cp->rx_len, DMA_FROM_DEVICE); |
1da177e4c Linux-2.6.12-rc2 |
659 660 661 |
pass_to_upper_layers(cp->rx_buf); make_and_setup_new_rx_buf(cp); } else { |
3f0fb4e85 Documentation/DMA... |
662 663 664 665 666 667 |
/* CPU should not write to * DMA_FROM_DEVICE-mapped area, * so dma_sync_single_for_device() is * not needed here. It would be required * for DMA_BIDIRECTIONAL mapping if * the memory was modified. |
1da177e4c Linux-2.6.12-rc2 |
668 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
669 670 671 672 |
give_rx_buf_to_card(cp); } } } |
77f2ea2f8 DMA-API: Clarify ... |
673 674 675 |
Drivers converted fully to this interface should not use virt_to_bus() any longer, nor should they use bus_to_virt(). Some drivers have to be changed a little bit, because there is no longer an equivalent to bus_to_virt() in the |
1da177e4c Linux-2.6.12-rc2 |
676 |
dynamic DMA mapping scheme - you have to always store the DMA addresses |
77f2ea2f8 DMA-API: Clarify ... |
677 678 |
returned by the dma_alloc_coherent(), dma_pool_alloc(), and dma_map_single() calls (dma_map_sg() stores them in the scatterlist itself if the platform |
1da177e4c Linux-2.6.12-rc2 |
679 680 |
supports dynamic DMA mapping in hardware) in your driver structures and/or in the card registers. |
216bf58f4 Documentation: co... |
681 682 |
All drivers should be using these interfaces with no exceptions. It is planned to completely remove virt_to_bus() and bus_to_virt() as |
1da177e4c Linux-2.6.12-rc2 |
683 684 |
they are entirely deprecated. Some ports already do not provide these as it is impossible to correctly support them. |
266921bdb DMA-API-HOWTO.txt... |
685 686 |
Handling Errors =============== |
4ae9ca825 Documentation: mo... |
687 688 689 |
DMA address space is limited on some architectures and an allocation failure can be determined by: |
77f2ea2f8 DMA-API: Clarify ... |
690 |
- checking if dma_alloc_coherent() returns NULL or dma_map_sg returns 0 |
4ae9ca825 Documentation: mo... |
691 |
|
77f2ea2f8 DMA-API: Clarify ... |
692 |
- checking the dma_addr_t returned from dma_map_single() and dma_map_page() |
266921bdb DMA-API-HOWTO.txt... |
693 |
by using dma_mapping_error():: |
4ae9ca825 Documentation: mo... |
694 695 696 697 698 699 700 701 702 703 |
dma_addr_t dma_handle; dma_handle = dma_map_single(dev, addr, size, direction); if (dma_mapping_error(dev, dma_handle)) { /* * reduce current DMA mapping usage, * delay and try again later or * reset driver. */ |
8d7f62e6a Documentation DMA... |
704 705 706 707 708 709 |
goto map_error_handling; } - unmap pages that are already mapped, when mapping error occurs in the middle of a multiple page mapping attempt. These example are applicable to dma_map_page() as well. |
266921bdb DMA-API-HOWTO.txt... |
710 |
Example 1:: |
8d7f62e6a Documentation DMA... |
711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 |
dma_addr_t dma_handle1; dma_addr_t dma_handle2; dma_handle1 = dma_map_single(dev, addr, size, direction); if (dma_mapping_error(dev, dma_handle1)) { /* * reduce current DMA mapping usage, * delay and try again later or * reset driver. */ goto map_error_handling1; } dma_handle2 = dma_map_single(dev, addr, size, direction); if (dma_mapping_error(dev, dma_handle2)) { /* * reduce current DMA mapping usage, * delay and try again later or * reset driver. */ goto map_error_handling2; } ... map_error_handling2: dma_unmap_single(dma_handle1); map_error_handling1: |
266921bdb DMA-API-HOWTO.txt... |
738 739 740 741 742 743 |
Example 2:: /* * if buffers are allocated in a loop, unmap all mapped buffers when * mapping error is detected in the middle */ |
8d7f62e6a Documentation DMA... |
744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 |
dma_addr_t dma_addr; dma_addr_t array[DMA_BUFFERS]; int save_index = 0; for (i = 0; i < DMA_BUFFERS; i++) { ... dma_addr = dma_map_single(dev, addr, size, direction); if (dma_mapping_error(dev, dma_addr)) { /* * reduce current DMA mapping usage, * delay and try again later or * reset driver. */ goto map_error_handling; } array[i].dma_addr = dma_addr; save_index++; } ... map_error_handling: for (i = 0; i < save_index; i++) { ... dma_unmap_single(array[i].dma_addr); |
4ae9ca825 Documentation: mo... |
775 |
} |
77f2ea2f8 DMA-API: Clarify ... |
776 |
Networking drivers must call dev_kfree_skb() to free the socket buffer |
4ae9ca825 Documentation: mo... |
777 778 779 780 781 782 783 |
and return NETDEV_TX_OK if the DMA mapping fails on the transmit hook (ndo_start_xmit). This means that the socket buffer is just dropped in the failure case. SCSI drivers must return SCSI_MLQUEUE_HOST_BUSY if the DMA mapping fails in the queuecommand hook. This means that the SCSI subsystem passes the command to the driver again later. |
266921bdb DMA-API-HOWTO.txt... |
784 785 |
Optimizing Unmap State Space Consumption ======================================== |
1da177e4c Linux-2.6.12-rc2 |
786 |
|
216bf58f4 Documentation: co... |
787 |
On many platforms, dma_unmap_{single,page}() is simply a nop. |
1da177e4c Linux-2.6.12-rc2 |
788 789 790 791 792 793 794 |
Therefore, keeping track of the mapping address and length is a waste of space. Instead of filling your drivers up with ifdefs and the like to "work around" this (which would defeat the whole purpose of a portable API) the following facilities are provided. Actually, instead of describing the macros one by one, we'll transform some example code. |
216bf58f4 Documentation: co... |
795 |
1) Use DEFINE_DMA_UNMAP_{ADDR,LEN} in state saving structures. |
266921bdb DMA-API-HOWTO.txt... |
796 |
Example, before:: |
1da177e4c Linux-2.6.12-rc2 |
797 798 799 800 801 802 |
struct ring_state { struct sk_buff *skb; dma_addr_t mapping; __u32 len; }; |
266921bdb DMA-API-HOWTO.txt... |
803 |
after:: |
1da177e4c Linux-2.6.12-rc2 |
804 805 806 |
struct ring_state { struct sk_buff *skb; |
216bf58f4 Documentation: co... |
807 808 |
DEFINE_DMA_UNMAP_ADDR(mapping); DEFINE_DMA_UNMAP_LEN(len); |
1da177e4c Linux-2.6.12-rc2 |
809 |
}; |
77f2ea2f8 DMA-API: Clarify ... |
810 |
2) Use dma_unmap_{addr,len}_set() to set these values. |
266921bdb DMA-API-HOWTO.txt... |
811 |
Example, before:: |
1da177e4c Linux-2.6.12-rc2 |
812 813 814 |
ringp->mapping = FOO; ringp->len = BAR; |
266921bdb DMA-API-HOWTO.txt... |
815 |
after:: |
1da177e4c Linux-2.6.12-rc2 |
816 |
|
216bf58f4 Documentation: co... |
817 818 |
dma_unmap_addr_set(ringp, mapping, FOO); dma_unmap_len_set(ringp, len, BAR); |
1da177e4c Linux-2.6.12-rc2 |
819 |
|
77f2ea2f8 DMA-API: Clarify ... |
820 |
3) Use dma_unmap_{addr,len}() to access these values. |
266921bdb DMA-API-HOWTO.txt... |
821 |
Example, before:: |
1da177e4c Linux-2.6.12-rc2 |
822 |
|
216bf58f4 Documentation: co... |
823 824 |
dma_unmap_single(dev, ringp->mapping, ringp->len, DMA_FROM_DEVICE); |
1da177e4c Linux-2.6.12-rc2 |
825 |
|
266921bdb DMA-API-HOWTO.txt... |
826 |
after:: |
1da177e4c Linux-2.6.12-rc2 |
827 |
|
216bf58f4 Documentation: co... |
828 829 830 831 |
dma_unmap_single(dev, dma_unmap_addr(ringp, mapping), dma_unmap_len(ringp, len), DMA_FROM_DEVICE); |
1da177e4c Linux-2.6.12-rc2 |
832 833 834 835 |
It really should be self-explanatory. We treat the ADDR and LEN separately, because it is possible for an implementation to only need the address in order to perform the unmap operation. |
266921bdb DMA-API-HOWTO.txt... |
836 837 |
Platform Issues =============== |
1da177e4c Linux-2.6.12-rc2 |
838 839 840 841 842 843 |
If you are just writing drivers for Linux and do not maintain an architecture port for the kernel, you can safely skip down to "Closing". 1) Struct scatterlist requirements. |
e92ae527e DMA-API-HOWTO: <a... |
844 845 |
You need to enable CONFIG_NEED_SG_DMA_LENGTH if the architecture supports IOMMUs (including software IOMMU). |
1da177e4c Linux-2.6.12-rc2 |
846 |
|
ce00f7feb Documentation: DM... |
847 |
2) ARCH_DMA_MINALIGN |
2fd74e25d Documentation/DMA... |
848 849 850 851 852 |
Architectures must ensure that kmalloc'ed buffer is DMA-safe. Drivers and subsystems depend on it. If an architecture isn't fully DMA-coherent (i.e. hardware doesn't ensure that data in the CPU cache is identical to data in main memory), |
ce00f7feb Documentation: DM... |
853 |
ARCH_DMA_MINALIGN must be set so that the memory allocator |
2fd74e25d Documentation/DMA... |
854 855 |
makes sure that kmalloc'ed buffer doesn't share a cache line with the others. See arch/arm/include/asm/cache.h as an example. |
ce00f7feb Documentation: DM... |
856 |
Note that ARCH_DMA_MINALIGN is about DMA memory alignment |
2fd74e25d Documentation/DMA... |
857 858 859 |
constraints. You don't need to worry about the architecture data alignment constraints (e.g. the alignment constraints about 64-bit objects). |
1da177e4c Linux-2.6.12-rc2 |
860 |
|
266921bdb DMA-API-HOWTO.txt... |
861 862 |
Closing ======= |
1da177e4c Linux-2.6.12-rc2 |
863 |
|
a33f32244 Documentation/: i... |
864 |
This document, and the API itself, would not be in its current |
1da177e4c Linux-2.6.12-rc2 |
865 866 |
form without the feedback and suggestions from numerous individuals. We would like to specifically mention, in no particular order, the |
266921bdb DMA-API-HOWTO.txt... |
867 |
following people:: |
1da177e4c Linux-2.6.12-rc2 |
868 869 870 871 872 873 874 875 |
Russell King <rmk@arm.linux.org.uk> Leo Dagum <dagum@barrel.engr.sgi.com> Ralf Baechle <ralf@oss.sgi.com> Grant Grundler <grundler@cup.hp.com> Jay Estabrook <Jay.Estabrook@compaq.com> Thomas Sailer <sailer@ife.ee.ethz.ch> Andrea Arcangeli <andrea@suse.de> |
26bbb29a2 Update Jens Axboe... |
876 |
Jens Axboe <jens.axboe@oracle.com> |
1da177e4c Linux-2.6.12-rc2 |
877 |
David Mosberger-Tang <davidm@hpl.hp.com> |