[linux.git] / Documentation / io-mapping.txt

The io_mapping functions in linux/io-mapping.h provide an abstraction for
efficiently mapping small regions of an I/O device to the CPU. The initial
usage is to support the large graphics aperture on 32-bit processors where
ioremap_wc cannot be used to statically map the entire aperture to the CPU
as it would consume too much of the kernel address space.

A mapping object is created during driver initialization using

	struct io_mapping *io_mapping_create_wc(unsigned long base,
						unsigned long size)

		'base' is the bus address of the region to be made
		mappable, while 'size' indicates how large a mapping region to
		enable. Both are in bytes.

		This _wc variant provides a mapping which may only be used
		with the io_mapping_map_atomic_wc or io_mapping_map_wc.

With this mapping object, individual pages can be mapped either atomically
or not, depending on the necessary scheduling environment. Of course, atomic
maps are more efficient:

	void *io_mapping_map_atomic_wc(struct io_mapping *mapping,
				       unsigned long offset)

		'offset' is the offset within the defined mapping region.
		Accessing addresses beyond the region specified in the
		creation function yields undefined results. Using an offset
		which is not page aligned yields an undefined result. The
		return value points to a single page in CPU address space.

		This _wc variant returns a write-combining map to the
		page and may only be used with mappings created by
		io_mapping_create_wc

		Note that the task may not sleep while holding this page
		mapped.

	void io_mapping_unmap_atomic(void *vaddr)

		'vaddr' must be the value returned by the last
		io_mapping_map_atomic_wc call. This unmaps the specified
		page and allows the task to sleep once again.

If you need to sleep while holding the lock, you can use the non-atomic
variant, although they may be significantly slower.

	void *io_mapping_map_wc(struct io_mapping *mapping,
				unsigned long offset)

		This works like io_mapping_map_atomic_wc except it allows
		the task to sleep while holding the page mapped.

	void io_mapping_unmap(void *vaddr)

		This works like io_mapping_unmap_atomic, except it is used
		for pages mapped with io_mapping_map_wc.

At driver close time, the io_mapping object must be freed:

	void io_mapping_free(struct io_mapping *mapping)

Current Implementation:

The initial implementation of these functions uses existing mapping
mechanisms and so provides only an abstraction layer and no new
functionality.

On 64-bit processors, io_mapping_create_wc calls ioremap_wc for the whole
range, creating a permanent kernel-visible mapping to the resource. The
map_atomic and map functions add the requested offset to the base of the
virtual address returned by ioremap_wc.

On 32-bit processors with HIGHMEM defined, io_mapping_map_atomic_wc uses
kmap_atomic_pfn to map the specified page in an atomic fashion;
kmap_atomic_pfn isn't really supposed to be used with device pages, but it
provides an efficient mapping for this usage.

On 32-bit processors without HIGHMEM defined, io_mapping_map_atomic_wc and
io_mapping_map_wc both use ioremap_wc, a terribly inefficient function which
performs an IPI to inform all processors about the new mapping. This results
in a significant performance penalty.
Commit	Line	Data
9663f2e6 KP	1	The io_mapping functions in linux/io-mapping.h provide an abstraction for
	2	efficiently mapping small regions of an I/O device to the CPU. The initial
	3	usage is to support the large graphics aperture on 32-bit processors where
	4	ioremap_wc cannot be used to statically map the entire aperture to the CPU
	5	as it would consume too much of the kernel address space.
	6
	7	A mapping object is created during driver initialization using
	8
	9	struct io_mapping *io_mapping_create_wc(unsigned long base,
	10	unsigned long size)
	11
	12	'base' is the bus address of the region to be made
	13	mappable, while 'size' indicates how large a mapping region to
	14	enable. Both are in bytes.
	15
	16	This _wc variant provides a mapping which may only be used
	17	with the io_mapping_map_atomic_wc or io_mapping_map_wc.
	18
	19	With this mapping object, individual pages can be mapped either atomically
	20	or not, depending on the necessary scheduling environment. Of course, atomic
	21	maps are more efficient:
	22
	23	void io_mapping_map_atomic_wc(struct io_mapping mapping,
	24	unsigned long offset)
	25
	26	'offset' is the offset within the defined mapping region.
	27	Accessing addresses beyond the region specified in the
	28	creation function yields undefined results. Using an offset
	29	which is not page aligned yields an undefined result. The
	30	return value points to a single page in CPU address space.
	31
	32	This _wc variant returns a write-combining map to the
	33	page and may only be used with mappings created by
	34	io_mapping_create_wc
	35
	36	Note that the task may not sleep while holding this page
	37	mapped.
	38
	39	void io_mapping_unmap_atomic(void *vaddr)
	40
d9a6ed1f	41	'vaddr' must be the value returned by the last
9663f2e6 KP	42	io_mapping_map_atomic_wc call. This unmaps the specified
	43	page and allows the task to sleep once again.
	44
	45	If you need to sleep while holding the lock, you can use the non-atomic
	46	variant, although they may be significantly slower.
	47
	48	void io_mapping_map_wc(struct io_mapping mapping,
	49	unsigned long offset)
	50
	51	This works like io_mapping_map_atomic_wc except it allows
	52	the task to sleep while holding the page mapped.
	53
	54	void io_mapping_unmap(void *vaddr)
	55
	56	This works like io_mapping_unmap_atomic, except it is used
	57	for pages mapped with io_mapping_map_wc.
	58
	59	At driver close time, the io_mapping object must be freed:
	60
	61	void io_mapping_free(struct io_mapping *mapping)
	62
	63	Current Implementation:
	64
	65	The initial implementation of these functions uses existing mapping
	66	mechanisms and so provides only an abstraction layer and no new
	67	functionality.
	68
	69	On 64-bit processors, io_mapping_create_wc calls ioremap_wc for the whole
	70	range, creating a permanent kernel-visible mapping to the resource. The
	71	map_atomic and map functions add the requested offset to the base of the
	72	virtual address returned by ioremap_wc.
	73
8d5c6603 KP	74	On 32-bit processors with HIGHMEM defined, io_mapping_map_atomic_wc uses
	75	kmap_atomic_pfn to map the specified page in an atomic fashion;
	76	kmap_atomic_pfn isn't really supposed to be used with device pages, but it
	77	provides an efficient mapping for this usage.
	78
	79	On 32-bit processors without HIGHMEM defined, io_mapping_map_atomic_wc and
	80	io_mapping_map_wc both use ioremap_wc, a terribly inefficient function which
	81	performs an IPI to inform all processors about the new mapping. This results
	82	in a significant performance penalty.