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