Loading...
1 Dynamic DMA mapping using the generic device
2 ============================================
3
4 James E.J. Bottomley <James.Bottomley@HansenPartnership.com>
5
6This document describes the DMA API. For a more gentle introduction
7of the API (and actual examples) see
8Documentation/DMA-API-HOWTO.txt.
9
10This API is split into two pieces. Part I describes the API. Part II
11describes the extensions to the API for supporting non-consistent
12memory machines. Unless you know that your driver absolutely has to
13support non-consistent platforms (this is usually only legacy
14platforms) you should only use the API described in part I.
15
16Part I - dma_ API
17-------------------------------------
18
19To get the dma_ API, you must #include <linux/dma-mapping.h>
20
21
22Part Ia - Using large dma-coherent buffers
23------------------------------------------
24
25void *
26dma_alloc_coherent(struct device *dev, size_t size,
27 dma_addr_t *dma_handle, gfp_t flag)
28
29Consistent memory is memory for which a write by either the device or
30the processor can immediately be read by the processor or device
31without having to worry about caching effects. (You may however need
32to make sure to flush the processor's write buffers before telling
33devices to read that memory.)
34
35This routine allocates a region of <size> bytes of consistent memory.
36It also returns a <dma_handle> which may be cast to an unsigned
37integer the same width as the bus and used as the physical address
38base of the region.
39
40Returns: a pointer to the allocated region (in the processor's virtual
41address space) or NULL if the allocation failed.
42
43Note: consistent memory can be expensive on some platforms, and the
44minimum allocation length may be as big as a page, so you should
45consolidate your requests for consistent memory as much as possible.
46The simplest way to do that is to use the dma_pool calls (see below).
47
48The flag parameter (dma_alloc_coherent only) allows the caller to
49specify the GFP_ flags (see kmalloc) for the allocation (the
50implementation may choose to ignore flags that affect the location of
51the returned memory, like GFP_DMA).
52
53void
54dma_free_coherent(struct device *dev, size_t size, void *cpu_addr,
55 dma_addr_t dma_handle)
56
57Free the region of consistent memory you previously allocated. dev,
58size and dma_handle must all be the same as those passed into the
59consistent allocate. cpu_addr must be the virtual address returned by
60the consistent allocate.
61
62Note that unlike their sibling allocation calls, these routines
63may only be called with IRQs enabled.
64
65
66Part Ib - Using small dma-coherent buffers
67------------------------------------------
68
69To get this part of the dma_ API, you must #include <linux/dmapool.h>
70
71Many drivers need lots of small dma-coherent memory regions for DMA
72descriptors or I/O buffers. Rather than allocating in units of a page
73or more using dma_alloc_coherent(), you can use DMA pools. These work
74much like a struct kmem_cache, except that they use the dma-coherent allocator,
75not __get_free_pages(). Also, they understand common hardware constraints
76for alignment, like queue heads needing to be aligned on N-byte boundaries.
77
78
79 struct dma_pool *
80 dma_pool_create(const char *name, struct device *dev,
81 size_t size, size_t align, size_t alloc);
82
83The pool create() routines initialize a pool of dma-coherent buffers
84for use with a given device. It must be called in a context which
85can sleep.
86
87The "name" is for diagnostics (like a struct kmem_cache name); dev and size
88are like what you'd pass to dma_alloc_coherent(). The device's hardware
89alignment requirement for this type of data is "align" (which is expressed
90in bytes, and must be a power of two). If your device has no boundary
91crossing restrictions, pass 0 for alloc; passing 4096 says memory allocated
92from this pool must not cross 4KByte boundaries.
93
94
95 void *dma_pool_alloc(struct dma_pool *pool, gfp_t gfp_flags,
96 dma_addr_t *dma_handle);
97
98This allocates memory from the pool; the returned memory will meet the size
99and alignment requirements specified at creation time. Pass GFP_ATOMIC to
100prevent blocking, or if it's permitted (not in_interrupt, not holding SMP locks),
101pass GFP_KERNEL to allow blocking. Like dma_alloc_coherent(), this returns
102two values: an address usable by the cpu, and the dma address usable by the
103pool's device.
104
105
106 void dma_pool_free(struct dma_pool *pool, void *vaddr,
107 dma_addr_t addr);
108
109This puts memory back into the pool. The pool is what was passed to
110the pool allocation routine; the cpu (vaddr) and dma addresses are what
111were returned when that routine allocated the memory being freed.
112
113
114 void dma_pool_destroy(struct dma_pool *pool);
115
116The pool destroy() routines free the resources of the pool. They must be
117called in a context which can sleep. Make sure you've freed all allocated
118memory back to the pool before you destroy it.
119
120
121Part Ic - DMA addressing limitations
122------------------------------------
123
124int
125dma_supported(struct device *dev, u64 mask)
126
127Checks to see if the device can support DMA to the memory described by
128mask.
129
130Returns: 1 if it can and 0 if it can't.
131
132Notes: This routine merely tests to see if the mask is possible. It
133won't change the current mask settings. It is more intended as an
134internal API for use by the platform than an external API for use by
135driver writers.
136
137int
138dma_set_mask(struct device *dev, u64 mask)
139
140Checks to see if the mask is possible and updates the device
141parameters if it is.
142
143Returns: 0 if successful and a negative error if not.
144
145int
146dma_set_coherent_mask(struct device *dev, u64 mask)
147
148Checks to see if the mask is possible and updates the device
149parameters if it is.
150
151Returns: 0 if successful and a negative error if not.
152
153u64
154dma_get_required_mask(struct device *dev)
155
156This API returns the mask that the platform requires to
157operate efficiently. Usually this means the returned mask
158is the minimum required to cover all of memory. Examining the
159required mask gives drivers with variable descriptor sizes the
160opportunity to use smaller descriptors as necessary.
161
162Requesting the required mask does not alter the current mask. If you
163wish to take advantage of it, you should issue a dma_set_mask()
164call to set the mask to the value returned.
165
166
167Part Id - Streaming DMA mappings
168--------------------------------
169
170dma_addr_t
171dma_map_single(struct device *dev, void *cpu_addr, size_t size,
172 enum dma_data_direction direction)
173
174Maps a piece of processor virtual memory so it can be accessed by the
175device and returns the physical handle of the memory.
176
177The direction for both api's may be converted freely by casting.
178However the dma_ API uses a strongly typed enumerator for its
179direction:
180
181DMA_NONE no direction (used for debugging)
182DMA_TO_DEVICE data is going from the memory to the device
183DMA_FROM_DEVICE data is coming from the device to the memory
184DMA_BIDIRECTIONAL direction isn't known
185
186Notes: Not all memory regions in a machine can be mapped by this
187API. Further, regions that appear to be physically contiguous in
188kernel virtual space may not be contiguous as physical memory. Since
189this API does not provide any scatter/gather capability, it will fail
190if the user tries to map a non-physically contiguous piece of memory.
191For this reason, it is recommended that memory mapped by this API be
192obtained only from sources which guarantee it to be physically contiguous
193(like kmalloc).
194
195Further, the physical address of the memory must be within the
196dma_mask of the device (the dma_mask represents a bit mask of the
197addressable region for the device. I.e., if the physical address of
198the memory anded with the dma_mask is still equal to the physical
199address, then the device can perform DMA to the memory). In order to
200ensure that the memory allocated by kmalloc is within the dma_mask,
201the driver may specify various platform-dependent flags to restrict
202the physical memory range of the allocation (e.g. on x86, GFP_DMA
203guarantees to be within the first 16Mb of available physical memory,
204as required by ISA devices).
205
206Note also that the above constraints on physical contiguity and
207dma_mask may not apply if the platform has an IOMMU (a device which
208supplies a physical to virtual mapping between the I/O memory bus and
209the device). However, to be portable, device driver writers may *not*
210assume that such an IOMMU exists.
211
212Warnings: Memory coherency operates at a granularity called the cache
213line width. In order for memory mapped by this API to operate
214correctly, the mapped region must begin exactly on a cache line
215boundary and end exactly on one (to prevent two separately mapped
216regions from sharing a single cache line). Since the cache line size
217may not be known at compile time, the API will not enforce this
218requirement. Therefore, it is recommended that driver writers who
219don't take special care to determine the cache line size at run time
220only map virtual regions that begin and end on page boundaries (which
221are guaranteed also to be cache line boundaries).
222
223DMA_TO_DEVICE synchronisation must be done after the last modification
224of the memory region by the software and before it is handed off to
225the driver. Once this primitive is used, memory covered by this
226primitive should be treated as read-only by the device. If the device
227may write to it at any point, it should be DMA_BIDIRECTIONAL (see
228below).
229
230DMA_FROM_DEVICE synchronisation must be done before the driver
231accesses data that may be changed by the device. This memory should
232be treated as read-only by the driver. If the driver needs to write
233to it at any point, it should be DMA_BIDIRECTIONAL (see below).
234
235DMA_BIDIRECTIONAL requires special handling: it means that the driver
236isn't sure if the memory was modified before being handed off to the
237device and also isn't sure if the device will also modify it. Thus,
238you must always sync bidirectional memory twice: once before the
239memory is handed off to the device (to make sure all memory changes
240are flushed from the processor) and once before the data may be
241accessed after being used by the device (to make sure any processor
242cache lines are updated with data that the device may have changed).
243
244void
245dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
246 enum dma_data_direction direction)
247
248Unmaps the region previously mapped. All the parameters passed in
249must be identical to those passed in (and returned) by the mapping
250API.
251
252dma_addr_t
253dma_map_page(struct device *dev, struct page *page,
254 unsigned long offset, size_t size,
255 enum dma_data_direction direction)
256void
257dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
258 enum dma_data_direction direction)
259
260API for mapping and unmapping for pages. All the notes and warnings
261for the other mapping APIs apply here. Also, although the <offset>
262and <size> parameters are provided to do partial page mapping, it is
263recommended that you never use these unless you really know what the
264cache width is.
265
266int
267dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
268
269In some circumstances dma_map_single and dma_map_page will fail to create
270a mapping. A driver can check for these errors by testing the returned
271dma address with dma_mapping_error(). A non-zero return value means the mapping
272could not be created and the driver should take appropriate action (e.g.
273reduce current DMA mapping usage or delay and try again later).
274
275 int
276 dma_map_sg(struct device *dev, struct scatterlist *sg,
277 int nents, enum dma_data_direction direction)
278
279Returns: the number of physical segments mapped (this may be shorter
280than <nents> passed in if some elements of the scatter/gather list are
281physically or virtually adjacent and an IOMMU maps them with a single
282entry).
283
284Please note that the sg cannot be mapped again if it has been mapped once.
285The mapping process is allowed to destroy information in the sg.
286
287As with the other mapping interfaces, dma_map_sg can fail. When it
288does, 0 is returned and a driver must take appropriate action. It is
289critical that the driver do something, in the case of a block driver
290aborting the request or even oopsing is better than doing nothing and
291corrupting the filesystem.
292
293With scatterlists, you use the resulting mapping like this:
294
295 int i, count = dma_map_sg(dev, sglist, nents, direction);
296 struct scatterlist *sg;
297
298 for_each_sg(sglist, sg, count, i) {
299 hw_address[i] = sg_dma_address(sg);
300 hw_len[i] = sg_dma_len(sg);
301 }
302
303where nents is the number of entries in the sglist.
304
305The implementation is free to merge several consecutive sglist entries
306into one (e.g. with an IOMMU, or if several pages just happen to be
307physically contiguous) and returns the actual number of sg entries it
308mapped them to. On failure 0, is returned.
309
310Then you should loop count times (note: this can be less than nents times)
311and use sg_dma_address() and sg_dma_len() macros where you previously
312accessed sg->address and sg->length as shown above.
313
314 void
315 dma_unmap_sg(struct device *dev, struct scatterlist *sg,
316 int nhwentries, enum dma_data_direction direction)
317
318Unmap the previously mapped scatter/gather list. All the parameters
319must be the same as those and passed in to the scatter/gather mapping
320API.
321
322Note: <nents> must be the number you passed in, *not* the number of
323physical entries returned.
324
325void
326dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
327 enum dma_data_direction direction)
328void
329dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size,
330 enum dma_data_direction direction)
331void
332dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
333 enum dma_data_direction direction)
334void
335dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
336 enum dma_data_direction direction)
337
338Synchronise a single contiguous or scatter/gather mapping for the cpu
339and device. With the sync_sg API, all the parameters must be the same
340as those passed into the single mapping API. With the sync_single API,
341you can use dma_handle and size parameters that aren't identical to
342those passed into the single mapping API to do a partial sync.
343
344Notes: You must do this:
345
346- Before reading values that have been written by DMA from the device
347 (use the DMA_FROM_DEVICE direction)
348- After writing values that will be written to the device using DMA
349 (use the DMA_TO_DEVICE) direction
350- before *and* after handing memory to the device if the memory is
351 DMA_BIDIRECTIONAL
352
353See also dma_map_single().
354
355dma_addr_t
356dma_map_single_attrs(struct device *dev, void *cpu_addr, size_t size,
357 enum dma_data_direction dir,
358 struct dma_attrs *attrs)
359
360void
361dma_unmap_single_attrs(struct device *dev, dma_addr_t dma_addr,
362 size_t size, enum dma_data_direction dir,
363 struct dma_attrs *attrs)
364
365int
366dma_map_sg_attrs(struct device *dev, struct scatterlist *sgl,
367 int nents, enum dma_data_direction dir,
368 struct dma_attrs *attrs)
369
370void
371dma_unmap_sg_attrs(struct device *dev, struct scatterlist *sgl,
372 int nents, enum dma_data_direction dir,
373 struct dma_attrs *attrs)
374
375The four functions above are just like the counterpart functions
376without the _attrs suffixes, except that they pass an optional
377struct dma_attrs*.
378
379struct dma_attrs encapsulates a set of "dma attributes". For the
380definition of struct dma_attrs see linux/dma-attrs.h.
381
382The interpretation of dma attributes is architecture-specific, and
383each attribute should be documented in Documentation/DMA-attributes.txt.
384
385If struct dma_attrs* is NULL, the semantics of each of these
386functions is identical to those of the corresponding function
387without the _attrs suffix. As a result dma_map_single_attrs()
388can generally replace dma_map_single(), etc.
389
390As an example of the use of the *_attrs functions, here's how
391you could pass an attribute DMA_ATTR_FOO when mapping memory
392for DMA:
393
394#include <linux/dma-attrs.h>
395/* DMA_ATTR_FOO should be defined in linux/dma-attrs.h and
396 * documented in Documentation/DMA-attributes.txt */
397...
398
399 DEFINE_DMA_ATTRS(attrs);
400 dma_set_attr(DMA_ATTR_FOO, &attrs);
401 ....
402 n = dma_map_sg_attrs(dev, sg, nents, DMA_TO_DEVICE, &attr);
403 ....
404
405Architectures that care about DMA_ATTR_FOO would check for its
406presence in their implementations of the mapping and unmapping
407routines, e.g.:
408
409void whizco_dma_map_sg_attrs(struct device *dev, dma_addr_t dma_addr,
410 size_t size, enum dma_data_direction dir,
411 struct dma_attrs *attrs)
412{
413 ....
414 int foo = dma_get_attr(DMA_ATTR_FOO, attrs);
415 ....
416 if (foo)
417 /* twizzle the frobnozzle */
418 ....
419
420
421Part II - Advanced dma_ usage
422-----------------------------
423
424Warning: These pieces of the DMA API should not be used in the
425majority of cases, since they cater for unlikely corner cases that
426don't belong in usual drivers.
427
428If you don't understand how cache line coherency works between a
429processor and an I/O device, you should not be using this part of the
430API at all.
431
432void *
433dma_alloc_noncoherent(struct device *dev, size_t size,
434 dma_addr_t *dma_handle, gfp_t flag)
435
436Identical to dma_alloc_coherent() except that the platform will
437choose to return either consistent or non-consistent memory as it sees
438fit. By using this API, you are guaranteeing to the platform that you
439have all the correct and necessary sync points for this memory in the
440driver should it choose to return non-consistent memory.
441
442Note: where the platform can return consistent memory, it will
443guarantee that the sync points become nops.
444
445Warning: Handling non-consistent memory is a real pain. You should
446only ever use this API if you positively know your driver will be
447required to work on one of the rare (usually non-PCI) architectures
448that simply cannot make consistent memory.
449
450void
451dma_free_noncoherent(struct device *dev, size_t size, void *cpu_addr,
452 dma_addr_t dma_handle)
453
454Free memory allocated by the nonconsistent API. All parameters must
455be identical to those passed in (and returned by
456dma_alloc_noncoherent()).
457
458int
459dma_get_cache_alignment(void)
460
461Returns the processor cache alignment. This is the absolute minimum
462alignment *and* width that you must observe when either mapping
463memory or doing partial flushes.
464
465Notes: This API may return a number *larger* than the actual cache
466line, but it will guarantee that one or more cache lines fit exactly
467into the width returned by this call. It will also always be a power
468of two for easy alignment.
469
470void
471dma_cache_sync(struct device *dev, void *vaddr, size_t size,
472 enum dma_data_direction direction)
473
474Do a partial sync of memory that was allocated by
475dma_alloc_noncoherent(), starting at virtual address vaddr and
476continuing on for size. Again, you *must* observe the cache line
477boundaries when doing this.
478
479int
480dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
481 dma_addr_t device_addr, size_t size, int
482 flags)
483
484Declare region of memory to be handed out by dma_alloc_coherent when
485it's asked for coherent memory for this device.
486
487bus_addr is the physical address to which the memory is currently
488assigned in the bus responding region (this will be used by the
489platform to perform the mapping).
490
491device_addr is the physical address the device needs to be programmed
492with actually to address this memory (this will be handed out as the
493dma_addr_t in dma_alloc_coherent()).
494
495size is the size of the area (must be multiples of PAGE_SIZE).
496
497flags can be or'd together and are:
498
499DMA_MEMORY_MAP - request that the memory returned from
500dma_alloc_coherent() be directly writable.
501
502DMA_MEMORY_IO - request that the memory returned from
503dma_alloc_coherent() be addressable using read/write/memcpy_toio etc.
504
505One or both of these flags must be present.
506
507DMA_MEMORY_INCLUDES_CHILDREN - make the declared memory be allocated by
508dma_alloc_coherent of any child devices of this one (for memory residing
509on a bridge).
510
511DMA_MEMORY_EXCLUSIVE - only allocate memory from the declared regions.
512Do not allow dma_alloc_coherent() to fall back to system memory when
513it's out of memory in the declared region.
514
515The return value will be either DMA_MEMORY_MAP or DMA_MEMORY_IO and
516must correspond to a passed in flag (i.e. no returning DMA_MEMORY_IO
517if only DMA_MEMORY_MAP were passed in) for success or zero for
518failure.
519
520Note, for DMA_MEMORY_IO returns, all subsequent memory returned by
521dma_alloc_coherent() may no longer be accessed directly, but instead
522must be accessed using the correct bus functions. If your driver
523isn't prepared to handle this contingency, it should not specify
524DMA_MEMORY_IO in the input flags.
525
526As a simplification for the platforms, only *one* such region of
527memory may be declared per device.
528
529For reasons of efficiency, most platforms choose to track the declared
530region only at the granularity of a page. For smaller allocations,
531you should use the dma_pool() API.
532
533void
534dma_release_declared_memory(struct device *dev)
535
536Remove the memory region previously declared from the system. This
537API performs *no* in-use checking for this region and will return
538unconditionally having removed all the required structures. It is the
539driver's job to ensure that no parts of this memory region are
540currently in use.
541
542void *
543dma_mark_declared_memory_occupied(struct device *dev,
544 dma_addr_t device_addr, size_t size)
545
546This is used to occupy specific regions of the declared space
547(dma_alloc_coherent() will hand out the first free region it finds).
548
549device_addr is the *device* address of the region requested.
550
551size is the size (and should be a page-sized multiple).
552
553The return value will be either a pointer to the processor virtual
554address of the memory, or an error (via PTR_ERR()) if any part of the
555region is occupied.
556
557Part III - Debug drivers use of the DMA-API
558-------------------------------------------
559
560The DMA-API as described above as some constraints. DMA addresses must be
561released with the corresponding function with the same size for example. With
562the advent of hardware IOMMUs it becomes more and more important that drivers
563do not violate those constraints. In the worst case such a violation can
564result in data corruption up to destroyed filesystems.
565
566To debug drivers and find bugs in the usage of the DMA-API checking code can
567be compiled into the kernel which will tell the developer about those
568violations. If your architecture supports it you can select the "Enable
569debugging of DMA-API usage" option in your kernel configuration. Enabling this
570option has a performance impact. Do not enable it in production kernels.
571
572If you boot the resulting kernel will contain code which does some bookkeeping
573about what DMA memory was allocated for which device. If this code detects an
574error it prints a warning message with some details into your kernel log. An
575example warning message may look like this:
576
577------------[ cut here ]------------
578WARNING: at /data2/repos/linux-2.6-iommu/lib/dma-debug.c:448
579 check_unmap+0x203/0x490()
580Hardware name:
581forcedeth 0000:00:08.0: DMA-API: device driver frees DMA memory with wrong
582 function [device address=0x00000000640444be] [size=66 bytes] [mapped as
583single] [unmapped as page]
584Modules linked in: nfsd exportfs bridge stp llc r8169
585Pid: 0, comm: swapper Tainted: G W 2.6.28-dmatest-09289-g8bb99c0 #1
586Call Trace:
587 <IRQ> [<ffffffff80240b22>] warn_slowpath+0xf2/0x130
588 [<ffffffff80647b70>] _spin_unlock+0x10/0x30
589 [<ffffffff80537e75>] usb_hcd_link_urb_to_ep+0x75/0xc0
590 [<ffffffff80647c22>] _spin_unlock_irqrestore+0x12/0x40
591 [<ffffffff8055347f>] ohci_urb_enqueue+0x19f/0x7c0
592 [<ffffffff80252f96>] queue_work+0x56/0x60
593 [<ffffffff80237e10>] enqueue_task_fair+0x20/0x50
594 [<ffffffff80539279>] usb_hcd_submit_urb+0x379/0xbc0
595 [<ffffffff803b78c3>] cpumask_next_and+0x23/0x40
596 [<ffffffff80235177>] find_busiest_group+0x207/0x8a0
597 [<ffffffff8064784f>] _spin_lock_irqsave+0x1f/0x50
598 [<ffffffff803c7ea3>] check_unmap+0x203/0x490
599 [<ffffffff803c8259>] debug_dma_unmap_page+0x49/0x50
600 [<ffffffff80485f26>] nv_tx_done_optimized+0xc6/0x2c0
601 [<ffffffff80486c13>] nv_nic_irq_optimized+0x73/0x2b0
602 [<ffffffff8026df84>] handle_IRQ_event+0x34/0x70
603 [<ffffffff8026ffe9>] handle_edge_irq+0xc9/0x150
604 [<ffffffff8020e3ab>] do_IRQ+0xcb/0x1c0
605 [<ffffffff8020c093>] ret_from_intr+0x0/0xa
606 <EOI> <4>---[ end trace f6435a98e2a38c0e ]---
607
608The driver developer can find the driver and the device including a stacktrace
609of the DMA-API call which caused this warning.
610
611Per default only the first error will result in a warning message. All other
612errors will only silently counted. This limitation exist to prevent the code
613from flooding your kernel log. To support debugging a device driver this can
614be disabled via debugfs. See the debugfs interface documentation below for
615details.
616
617The debugfs directory for the DMA-API debugging code is called dma-api/. In
618this directory the following files can currently be found:
619
620 dma-api/all_errors This file contains a numeric value. If this
621 value is not equal to zero the debugging code
622 will print a warning for every error it finds
623 into the kernel log. Be careful with this
624 option, as it can easily flood your logs.
625
626 dma-api/disabled This read-only file contains the character 'Y'
627 if the debugging code is disabled. This can
628 happen when it runs out of memory or if it was
629 disabled at boot time
630
631 dma-api/error_count This file is read-only and shows the total
632 numbers of errors found.
633
634 dma-api/num_errors The number in this file shows how many
635 warnings will be printed to the kernel log
636 before it stops. This number is initialized to
637 one at system boot and be set by writing into
638 this file
639
640 dma-api/min_free_entries
641 This read-only file can be read to get the
642 minimum number of free dma_debug_entries the
643 allocator has ever seen. If this value goes
644 down to zero the code will disable itself
645 because it is not longer reliable.
646
647 dma-api/num_free_entries
648 The current number of free dma_debug_entries
649 in the allocator.
650
651 dma-api/driver-filter
652 You can write a name of a driver into this file
653 to limit the debug output to requests from that
654 particular driver. Write an empty string to
655 that file to disable the filter and see
656 all errors again.
657
658If you have this code compiled into your kernel it will be enabled by default.
659If you want to boot without the bookkeeping anyway you can provide
660'dma_debug=off' as a boot parameter. This will disable DMA-API debugging.
661Notice that you can not enable it again at runtime. You have to reboot to do
662so.
663
664If you want to see debug messages only for a special device driver you can
665specify the dma_debug_driver=<drivername> parameter. This will enable the
666driver filter at boot time. The debug code will only print errors for that
667driver afterwards. This filter can be disabled or changed later using debugfs.
668
669When the code disables itself at runtime this is most likely because it ran
670out of dma_debug_entries. These entries are preallocated at boot. The number
671of preallocated entries is defined per architecture. If it is too low for you
672boot with 'dma_debug_entries=<your_desired_number>' to overwrite the
673architectural default.
1============================================
2Dynamic DMA mapping using the generic device
3============================================
4
5:Author: James E.J. Bottomley <James.Bottomley@HansenPartnership.com>
6
7This document describes the DMA API. For a more gentle introduction
8of the API (and actual examples), see Documentation/DMA-API-HOWTO.txt.
9
10This API is split into two pieces. Part I describes the basic API.
11Part II describes extensions for supporting non-consistent memory
12machines. Unless you know that your driver absolutely has to support
13non-consistent platforms (this is usually only legacy platforms) you
14should only use the API described in part I.
15
16Part I - dma_API
17----------------
18
19To get the dma_API, you must #include <linux/dma-mapping.h>. This
20provides dma_addr_t and the interfaces described below.
21
22A dma_addr_t can hold any valid DMA address for the platform. It can be
23given to a device to use as a DMA source or target. A CPU cannot reference
24a dma_addr_t directly because there may be translation between its physical
25address space and the DMA address space.
26
27Part Ia - Using large DMA-coherent buffers
28------------------------------------------
29
30::
31
32 void *
33 dma_alloc_coherent(struct device *dev, size_t size,
34 dma_addr_t *dma_handle, gfp_t flag)
35
36Consistent memory is memory for which a write by either the device or
37the processor can immediately be read by the processor or device
38without having to worry about caching effects. (You may however need
39to make sure to flush the processor's write buffers before telling
40devices to read that memory.)
41
42This routine allocates a region of <size> bytes of consistent memory.
43
44It returns a pointer to the allocated region (in the processor's virtual
45address space) or NULL if the allocation failed.
46
47It also returns a <dma_handle> which may be cast to an unsigned integer the
48same width as the bus and given to the device as the DMA address base of
49the region.
50
51Note: consistent memory can be expensive on some platforms, and the
52minimum allocation length may be as big as a page, so you should
53consolidate your requests for consistent memory as much as possible.
54The simplest way to do that is to use the dma_pool calls (see below).
55
56The flag parameter (dma_alloc_coherent() only) allows the caller to
57specify the ``GFP_`` flags (see kmalloc()) for the allocation (the
58implementation may choose to ignore flags that affect the location of
59the returned memory, like GFP_DMA).
60
61::
62
63 void
64 dma_free_coherent(struct device *dev, size_t size, void *cpu_addr,
65 dma_addr_t dma_handle)
66
67Free a region of consistent memory you previously allocated. dev,
68size and dma_handle must all be the same as those passed into
69dma_alloc_coherent(). cpu_addr must be the virtual address returned by
70the dma_alloc_coherent().
71
72Note that unlike their sibling allocation calls, these routines
73may only be called with IRQs enabled.
74
75
76Part Ib - Using small DMA-coherent buffers
77------------------------------------------
78
79To get this part of the dma_API, you must #include <linux/dmapool.h>
80
81Many drivers need lots of small DMA-coherent memory regions for DMA
82descriptors or I/O buffers. Rather than allocating in units of a page
83or more using dma_alloc_coherent(), you can use DMA pools. These work
84much like a struct kmem_cache, except that they use the DMA-coherent allocator,
85not __get_free_pages(). Also, they understand common hardware constraints
86for alignment, like queue heads needing to be aligned on N-byte boundaries.
87
88
89::
90
91 struct dma_pool *
92 dma_pool_create(const char *name, struct device *dev,
93 size_t size, size_t align, size_t alloc);
94
95dma_pool_create() initializes a pool of DMA-coherent buffers
96for use with a given device. It must be called in a context which
97can sleep.
98
99The "name" is for diagnostics (like a struct kmem_cache name); dev and size
100are like what you'd pass to dma_alloc_coherent(). The device's hardware
101alignment requirement for this type of data is "align" (which is expressed
102in bytes, and must be a power of two). If your device has no boundary
103crossing restrictions, pass 0 for alloc; passing 4096 says memory allocated
104from this pool must not cross 4KByte boundaries.
105
106::
107
108 void *
109 dma_pool_zalloc(struct dma_pool *pool, gfp_t mem_flags,
110 dma_addr_t *handle)
111
112Wraps dma_pool_alloc() and also zeroes the returned memory if the
113allocation attempt succeeded.
114
115
116::
117
118 void *
119 dma_pool_alloc(struct dma_pool *pool, gfp_t gfp_flags,
120 dma_addr_t *dma_handle);
121
122This allocates memory from the pool; the returned memory will meet the
123size and alignment requirements specified at creation time. Pass
124GFP_ATOMIC to prevent blocking, or if it's permitted (not
125in_interrupt, not holding SMP locks), pass GFP_KERNEL to allow
126blocking. Like dma_alloc_coherent(), this returns two values: an
127address usable by the CPU, and the DMA address usable by the pool's
128device.
129
130::
131
132 void
133 dma_pool_free(struct dma_pool *pool, void *vaddr,
134 dma_addr_t addr);
135
136This puts memory back into the pool. The pool is what was passed to
137dma_pool_alloc(); the CPU (vaddr) and DMA addresses are what
138were returned when that routine allocated the memory being freed.
139
140::
141
142 void
143 dma_pool_destroy(struct dma_pool *pool);
144
145dma_pool_destroy() frees the resources of the pool. It must be
146called in a context which can sleep. Make sure you've freed all allocated
147memory back to the pool before you destroy it.
148
149
150Part Ic - DMA addressing limitations
151------------------------------------
152
153::
154
155 int
156 dma_set_mask_and_coherent(struct device *dev, u64 mask)
157
158Checks to see if the mask is possible and updates the device
159streaming and coherent DMA mask parameters if it is.
160
161Returns: 0 if successful and a negative error if not.
162
163::
164
165 int
166 dma_set_mask(struct device *dev, u64 mask)
167
168Checks to see if the mask is possible and updates the device
169parameters if it is.
170
171Returns: 0 if successful and a negative error if not.
172
173::
174
175 int
176 dma_set_coherent_mask(struct device *dev, u64 mask)
177
178Checks to see if the mask is possible and updates the device
179parameters if it is.
180
181Returns: 0 if successful and a negative error if not.
182
183::
184
185 u64
186 dma_get_required_mask(struct device *dev)
187
188This API returns the mask that the platform requires to
189operate efficiently. Usually this means the returned mask
190is the minimum required to cover all of memory. Examining the
191required mask gives drivers with variable descriptor sizes the
192opportunity to use smaller descriptors as necessary.
193
194Requesting the required mask does not alter the current mask. If you
195wish to take advantage of it, you should issue a dma_set_mask()
196call to set the mask to the value returned.
197
198::
199
200 size_t
201 dma_max_mapping_size(struct device *dev);
202
203Returns the maximum size of a mapping for the device. The size parameter
204of the mapping functions like dma_map_single(), dma_map_page() and
205others should not be larger than the returned value.
206
207::
208
209 unsigned long
210 dma_get_merge_boundary(struct device *dev);
211
212Returns the DMA merge boundary. If the device cannot merge any the DMA address
213segments, the function returns 0.
214
215Part Id - Streaming DMA mappings
216--------------------------------
217
218::
219
220 dma_addr_t
221 dma_map_single(struct device *dev, void *cpu_addr, size_t size,
222 enum dma_data_direction direction)
223
224Maps a piece of processor virtual memory so it can be accessed by the
225device and returns the DMA address of the memory.
226
227The direction for both APIs may be converted freely by casting.
228However the dma_API uses a strongly typed enumerator for its
229direction:
230
231======================= =============================================
232DMA_NONE no direction (used for debugging)
233DMA_TO_DEVICE data is going from the memory to the device
234DMA_FROM_DEVICE data is coming from the device to the memory
235DMA_BIDIRECTIONAL direction isn't known
236======================= =============================================
237
238.. note::
239
240 Not all memory regions in a machine can be mapped by this API.
241 Further, contiguous kernel virtual space may not be contiguous as
242 physical memory. Since this API does not provide any scatter/gather
243 capability, it will fail if the user tries to map a non-physically
244 contiguous piece of memory. For this reason, memory to be mapped by
245 this API should be obtained from sources which guarantee it to be
246 physically contiguous (like kmalloc).
247
248 Further, the DMA address of the memory must be within the
249 dma_mask of the device (the dma_mask is a bit mask of the
250 addressable region for the device, i.e., if the DMA address of
251 the memory ANDed with the dma_mask is still equal to the DMA
252 address, then the device can perform DMA to the memory). To
253 ensure that the memory allocated by kmalloc is within the dma_mask,
254 the driver may specify various platform-dependent flags to restrict
255 the DMA address range of the allocation (e.g., on x86, GFP_DMA
256 guarantees to be within the first 16MB of available DMA addresses,
257 as required by ISA devices).
258
259 Note also that the above constraints on physical contiguity and
260 dma_mask may not apply if the platform has an IOMMU (a device which
261 maps an I/O DMA address to a physical memory address). However, to be
262 portable, device driver writers may *not* assume that such an IOMMU
263 exists.
264
265.. warning::
266
267 Memory coherency operates at a granularity called the cache
268 line width. In order for memory mapped by this API to operate
269 correctly, the mapped region must begin exactly on a cache line
270 boundary and end exactly on one (to prevent two separately mapped
271 regions from sharing a single cache line). Since the cache line size
272 may not be known at compile time, the API will not enforce this
273 requirement. Therefore, it is recommended that driver writers who
274 don't take special care to determine the cache line size at run time
275 only map virtual regions that begin and end on page boundaries (which
276 are guaranteed also to be cache line boundaries).
277
278 DMA_TO_DEVICE synchronisation must be done after the last modification
279 of the memory region by the software and before it is handed off to
280 the device. Once this primitive is used, memory covered by this
281 primitive should be treated as read-only by the device. If the device
282 may write to it at any point, it should be DMA_BIDIRECTIONAL (see
283 below).
284
285 DMA_FROM_DEVICE synchronisation must be done before the driver
286 accesses data that may be changed by the device. This memory should
287 be treated as read-only by the driver. If the driver needs to write
288 to it at any point, it should be DMA_BIDIRECTIONAL (see below).
289
290 DMA_BIDIRECTIONAL requires special handling: it means that the driver
291 isn't sure if the memory was modified before being handed off to the
292 device and also isn't sure if the device will also modify it. Thus,
293 you must always sync bidirectional memory twice: once before the
294 memory is handed off to the device (to make sure all memory changes
295 are flushed from the processor) and once before the data may be
296 accessed after being used by the device (to make sure any processor
297 cache lines are updated with data that the device may have changed).
298
299::
300
301 void
302 dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
303 enum dma_data_direction direction)
304
305Unmaps the region previously mapped. All the parameters passed in
306must be identical to those passed in (and returned) by the mapping
307API.
308
309::
310
311 dma_addr_t
312 dma_map_page(struct device *dev, struct page *page,
313 unsigned long offset, size_t size,
314 enum dma_data_direction direction)
315
316 void
317 dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
318 enum dma_data_direction direction)
319
320API for mapping and unmapping for pages. All the notes and warnings
321for the other mapping APIs apply here. Also, although the <offset>
322and <size> parameters are provided to do partial page mapping, it is
323recommended that you never use these unless you really know what the
324cache width is.
325
326::
327
328 dma_addr_t
329 dma_map_resource(struct device *dev, phys_addr_t phys_addr, size_t size,
330 enum dma_data_direction dir, unsigned long attrs)
331
332 void
333 dma_unmap_resource(struct device *dev, dma_addr_t addr, size_t size,
334 enum dma_data_direction dir, unsigned long attrs)
335
336API for mapping and unmapping for MMIO resources. All the notes and
337warnings for the other mapping APIs apply here. The API should only be
338used to map device MMIO resources, mapping of RAM is not permitted.
339
340::
341
342 int
343 dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
344
345In some circumstances dma_map_single(), dma_map_page() and dma_map_resource()
346will fail to create a mapping. A driver can check for these errors by testing
347the returned DMA address with dma_mapping_error(). A non-zero return value
348means the mapping could not be created and the driver should take appropriate
349action (e.g. reduce current DMA mapping usage or delay and try again later).
350
351::
352
353 int
354 dma_map_sg(struct device *dev, struct scatterlist *sg,
355 int nents, enum dma_data_direction direction)
356
357Returns: the number of DMA address segments mapped (this may be shorter
358than <nents> passed in if some elements of the scatter/gather list are
359physically or virtually adjacent and an IOMMU maps them with a single
360entry).
361
362Please note that the sg cannot be mapped again if it has been mapped once.
363The mapping process is allowed to destroy information in the sg.
364
365As with the other mapping interfaces, dma_map_sg() can fail. When it
366does, 0 is returned and a driver must take appropriate action. It is
367critical that the driver do something, in the case of a block driver
368aborting the request or even oopsing is better than doing nothing and
369corrupting the filesystem.
370
371With scatterlists, you use the resulting mapping like this::
372
373 int i, count = dma_map_sg(dev, sglist, nents, direction);
374 struct scatterlist *sg;
375
376 for_each_sg(sglist, sg, count, i) {
377 hw_address[i] = sg_dma_address(sg);
378 hw_len[i] = sg_dma_len(sg);
379 }
380
381where nents is the number of entries in the sglist.
382
383The implementation is free to merge several consecutive sglist entries
384into one (e.g. with an IOMMU, or if several pages just happen to be
385physically contiguous) and returns the actual number of sg entries it
386mapped them to. On failure 0, is returned.
387
388Then you should loop count times (note: this can be less than nents times)
389and use sg_dma_address() and sg_dma_len() macros where you previously
390accessed sg->address and sg->length as shown above.
391
392::
393
394 void
395 dma_unmap_sg(struct device *dev, struct scatterlist *sg,
396 int nents, enum dma_data_direction direction)
397
398Unmap the previously mapped scatter/gather list. All the parameters
399must be the same as those and passed in to the scatter/gather mapping
400API.
401
402Note: <nents> must be the number you passed in, *not* the number of
403DMA address entries returned.
404
405::
406
407 void
408 dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
409 size_t size,
410 enum dma_data_direction direction)
411
412 void
413 dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
414 size_t size,
415 enum dma_data_direction direction)
416
417 void
418 dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
419 int nents,
420 enum dma_data_direction direction)
421
422 void
423 dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
424 int nents,
425 enum dma_data_direction direction)
426
427Synchronise a single contiguous or scatter/gather mapping for the CPU
428and device. With the sync_sg API, all the parameters must be the same
429as those passed into the single mapping API. With the sync_single API,
430you can use dma_handle and size parameters that aren't identical to
431those passed into the single mapping API to do a partial sync.
432
433
434.. note::
435
436 You must do this:
437
438 - Before reading values that have been written by DMA from the device
439 (use the DMA_FROM_DEVICE direction)
440 - After writing values that will be written to the device using DMA
441 (use the DMA_TO_DEVICE) direction
442 - before *and* after handing memory to the device if the memory is
443 DMA_BIDIRECTIONAL
444
445See also dma_map_single().
446
447::
448
449 dma_addr_t
450 dma_map_single_attrs(struct device *dev, void *cpu_addr, size_t size,
451 enum dma_data_direction dir,
452 unsigned long attrs)
453
454 void
455 dma_unmap_single_attrs(struct device *dev, dma_addr_t dma_addr,
456 size_t size, enum dma_data_direction dir,
457 unsigned long attrs)
458
459 int
460 dma_map_sg_attrs(struct device *dev, struct scatterlist *sgl,
461 int nents, enum dma_data_direction dir,
462 unsigned long attrs)
463
464 void
465 dma_unmap_sg_attrs(struct device *dev, struct scatterlist *sgl,
466 int nents, enum dma_data_direction dir,
467 unsigned long attrs)
468
469The four functions above are just like the counterpart functions
470without the _attrs suffixes, except that they pass an optional
471dma_attrs.
472
473The interpretation of DMA attributes is architecture-specific, and
474each attribute should be documented in Documentation/DMA-attributes.txt.
475
476If dma_attrs are 0, the semantics of each of these functions
477is identical to those of the corresponding function
478without the _attrs suffix. As a result dma_map_single_attrs()
479can generally replace dma_map_single(), etc.
480
481As an example of the use of the ``*_attrs`` functions, here's how
482you could pass an attribute DMA_ATTR_FOO when mapping memory
483for DMA::
484
485 #include <linux/dma-mapping.h>
486 /* DMA_ATTR_FOO should be defined in linux/dma-mapping.h and
487 * documented in Documentation/DMA-attributes.txt */
488 ...
489
490 unsigned long attr;
491 attr |= DMA_ATTR_FOO;
492 ....
493 n = dma_map_sg_attrs(dev, sg, nents, DMA_TO_DEVICE, attr);
494 ....
495
496Architectures that care about DMA_ATTR_FOO would check for its
497presence in their implementations of the mapping and unmapping
498routines, e.g.:::
499
500 void whizco_dma_map_sg_attrs(struct device *dev, dma_addr_t dma_addr,
501 size_t size, enum dma_data_direction dir,
502 unsigned long attrs)
503 {
504 ....
505 if (attrs & DMA_ATTR_FOO)
506 /* twizzle the frobnozzle */
507 ....
508 }
509
510
511Part II - Advanced dma usage
512----------------------------
513
514Warning: These pieces of the DMA API should not be used in the
515majority of cases, since they cater for unlikely corner cases that
516don't belong in usual drivers.
517
518If you don't understand how cache line coherency works between a
519processor and an I/O device, you should not be using this part of the
520API at all.
521
522::
523
524 void *
525 dma_alloc_attrs(struct device *dev, size_t size, dma_addr_t *dma_handle,
526 gfp_t flag, unsigned long attrs)
527
528Identical to dma_alloc_coherent() except that when the
529DMA_ATTR_NON_CONSISTENT flags is passed in the attrs argument, the
530platform will choose to return either consistent or non-consistent memory
531as it sees fit. By using this API, you are guaranteeing to the platform
532that you have all the correct and necessary sync points for this memory
533in the driver should it choose to return non-consistent memory.
534
535Note: where the platform can return consistent memory, it will
536guarantee that the sync points become nops.
537
538Warning: Handling non-consistent memory is a real pain. You should
539only use this API if you positively know your driver will be
540required to work on one of the rare (usually non-PCI) architectures
541that simply cannot make consistent memory.
542
543::
544
545 void
546 dma_free_attrs(struct device *dev, size_t size, void *cpu_addr,
547 dma_addr_t dma_handle, unsigned long attrs)
548
549Free memory allocated by the dma_alloc_attrs(). All common
550parameters must be identical to those otherwise passed to dma_free_coherent,
551and the attrs argument must be identical to the attrs passed to
552dma_alloc_attrs().
553
554::
555
556 int
557 dma_get_cache_alignment(void)
558
559Returns the processor cache alignment. This is the absolute minimum
560alignment *and* width that you must observe when either mapping
561memory or doing partial flushes.
562
563.. note::
564
565 This API may return a number *larger* than the actual cache
566 line, but it will guarantee that one or more cache lines fit exactly
567 into the width returned by this call. It will also always be a power
568 of two for easy alignment.
569
570::
571
572 void
573 dma_cache_sync(struct device *dev, void *vaddr, size_t size,
574 enum dma_data_direction direction)
575
576Do a partial sync of memory that was allocated by dma_alloc_attrs() with
577the DMA_ATTR_NON_CONSISTENT flag starting at virtual address vaddr and
578continuing on for size. Again, you *must* observe the cache line
579boundaries when doing this.
580
581::
582
583 int
584 dma_declare_coherent_memory(struct device *dev, phys_addr_t phys_addr,
585 dma_addr_t device_addr, size_t size);
586
587Declare region of memory to be handed out by dma_alloc_coherent() when
588it's asked for coherent memory for this device.
589
590phys_addr is the CPU physical address to which the memory is currently
591assigned (this will be ioremapped so the CPU can access the region).
592
593device_addr is the DMA address the device needs to be programmed
594with to actually address this memory (this will be handed out as the
595dma_addr_t in dma_alloc_coherent()).
596
597size is the size of the area (must be multiples of PAGE_SIZE).
598
599As a simplification for the platforms, only *one* such region of
600memory may be declared per device.
601
602For reasons of efficiency, most platforms choose to track the declared
603region only at the granularity of a page. For smaller allocations,
604you should use the dma_pool() API.
605
606Part III - Debug drivers use of the DMA-API
607-------------------------------------------
608
609The DMA-API as described above has some constraints. DMA addresses must be
610released with the corresponding function with the same size for example. With
611the advent of hardware IOMMUs it becomes more and more important that drivers
612do not violate those constraints. In the worst case such a violation can
613result in data corruption up to destroyed filesystems.
614
615To debug drivers and find bugs in the usage of the DMA-API checking code can
616be compiled into the kernel which will tell the developer about those
617violations. If your architecture supports it you can select the "Enable
618debugging of DMA-API usage" option in your kernel configuration. Enabling this
619option has a performance impact. Do not enable it in production kernels.
620
621If you boot the resulting kernel will contain code which does some bookkeeping
622about what DMA memory was allocated for which device. If this code detects an
623error it prints a warning message with some details into your kernel log. An
624example warning message may look like this::
625
626 WARNING: at /data2/repos/linux-2.6-iommu/lib/dma-debug.c:448
627 check_unmap+0x203/0x490()
628 Hardware name:
629 forcedeth 0000:00:08.0: DMA-API: device driver frees DMA memory with wrong
630 function [device address=0x00000000640444be] [size=66 bytes] [mapped as
631 single] [unmapped as page]
632 Modules linked in: nfsd exportfs bridge stp llc r8169
633 Pid: 0, comm: swapper Tainted: G W 2.6.28-dmatest-09289-g8bb99c0 #1
634 Call Trace:
635 <IRQ> [<ffffffff80240b22>] warn_slowpath+0xf2/0x130
636 [<ffffffff80647b70>] _spin_unlock+0x10/0x30
637 [<ffffffff80537e75>] usb_hcd_link_urb_to_ep+0x75/0xc0
638 [<ffffffff80647c22>] _spin_unlock_irqrestore+0x12/0x40
639 [<ffffffff8055347f>] ohci_urb_enqueue+0x19f/0x7c0
640 [<ffffffff80252f96>] queue_work+0x56/0x60
641 [<ffffffff80237e10>] enqueue_task_fair+0x20/0x50
642 [<ffffffff80539279>] usb_hcd_submit_urb+0x379/0xbc0
643 [<ffffffff803b78c3>] cpumask_next_and+0x23/0x40
644 [<ffffffff80235177>] find_busiest_group+0x207/0x8a0
645 [<ffffffff8064784f>] _spin_lock_irqsave+0x1f/0x50
646 [<ffffffff803c7ea3>] check_unmap+0x203/0x490
647 [<ffffffff803c8259>] debug_dma_unmap_page+0x49/0x50
648 [<ffffffff80485f26>] nv_tx_done_optimized+0xc6/0x2c0
649 [<ffffffff80486c13>] nv_nic_irq_optimized+0x73/0x2b0
650 [<ffffffff8026df84>] handle_IRQ_event+0x34/0x70
651 [<ffffffff8026ffe9>] handle_edge_irq+0xc9/0x150
652 [<ffffffff8020e3ab>] do_IRQ+0xcb/0x1c0
653 [<ffffffff8020c093>] ret_from_intr+0x0/0xa
654 <EOI> <4>---[ end trace f6435a98e2a38c0e ]---
655
656The driver developer can find the driver and the device including a stacktrace
657of the DMA-API call which caused this warning.
658
659Per default only the first error will result in a warning message. All other
660errors will only silently counted. This limitation exist to prevent the code
661from flooding your kernel log. To support debugging a device driver this can
662be disabled via debugfs. See the debugfs interface documentation below for
663details.
664
665The debugfs directory for the DMA-API debugging code is called dma-api/. In
666this directory the following files can currently be found:
667
668=============================== ===============================================
669dma-api/all_errors This file contains a numeric value. If this
670 value is not equal to zero the debugging code
671 will print a warning for every error it finds
672 into the kernel log. Be careful with this
673 option, as it can easily flood your logs.
674
675dma-api/disabled This read-only file contains the character 'Y'
676 if the debugging code is disabled. This can
677 happen when it runs out of memory or if it was
678 disabled at boot time
679
680dma-api/dump This read-only file contains current DMA
681 mappings.
682
683dma-api/error_count This file is read-only and shows the total
684 numbers of errors found.
685
686dma-api/num_errors The number in this file shows how many
687 warnings will be printed to the kernel log
688 before it stops. This number is initialized to
689 one at system boot and be set by writing into
690 this file
691
692dma-api/min_free_entries This read-only file can be read to get the
693 minimum number of free dma_debug_entries the
694 allocator has ever seen. If this value goes
695 down to zero the code will attempt to increase
696 nr_total_entries to compensate.
697
698dma-api/num_free_entries The current number of free dma_debug_entries
699 in the allocator.
700
701dma-api/nr_total_entries The total number of dma_debug_entries in the
702 allocator, both free and used.
703
704dma-api/driver_filter You can write a name of a driver into this file
705 to limit the debug output to requests from that
706 particular driver. Write an empty string to
707 that file to disable the filter and see
708 all errors again.
709=============================== ===============================================
710
711If you have this code compiled into your kernel it will be enabled by default.
712If you want to boot without the bookkeeping anyway you can provide
713'dma_debug=off' as a boot parameter. This will disable DMA-API debugging.
714Notice that you can not enable it again at runtime. You have to reboot to do
715so.
716
717If you want to see debug messages only for a special device driver you can
718specify the dma_debug_driver=<drivername> parameter. This will enable the
719driver filter at boot time. The debug code will only print errors for that
720driver afterwards. This filter can be disabled or changed later using debugfs.
721
722When the code disables itself at runtime this is most likely because it ran
723out of dma_debug_entries and was unable to allocate more on-demand. 65536
724entries are preallocated at boot - if this is too low for you boot with
725'dma_debug_entries=<your_desired_number>' to overwrite the default. Note
726that the code allocates entries in batches, so the exact number of
727preallocated entries may be greater than the actual number requested. The
728code will print to the kernel log each time it has dynamically allocated
729as many entries as were initially preallocated. This is to indicate that a
730larger preallocation size may be appropriate, or if it happens continually
731that a driver may be leaking mappings.
732
733::
734
735 void
736 debug_dma_mapping_error(struct device *dev, dma_addr_t dma_addr);
737
738dma-debug interface debug_dma_mapping_error() to debug drivers that fail
739to check DMA mapping errors on addresses returned by dma_map_single() and
740dma_map_page() interfaces. This interface clears a flag set by
741debug_dma_map_page() to indicate that dma_mapping_error() has been called by
742the driver. When driver does unmap, debug_dma_unmap() checks the flag and if
743this flag is still set, prints warning message that includes call trace that
744leads up to the unmap. This interface can be called from dma_mapping_error()
745routines to enable DMA mapping error check debugging.