Move DMA-mapping.txt to Documentation/PCI/.
DMA-mapping.txt was supposed to be moved from Documentation/ to
Documentation/PCI/. The 00-INDEX files in those two directories
were updated, along with a few other text files, but the file
itself somehow escaped being moved, so move it and update more
text files and source files with its new location.
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This document describes the DMA API. For a more gentle introduction
phrased in terms of the pci_ equivalents (and actual examples) see
-DMA-mapping.txt
+Documentation/PCI/PCI-DMA-mapping.txt.
This API is split into two pieces. Part I describes the API and the
corresponding pci_ API. Part II describes the extensions to the API
[ NOTE: The virt_to_bus() and bus_to_virt() functions have been
- superseded by the functionality provided by the PCI DMA
- interface (see Documentation/DMA-mapping.txt). They continue
+ superseded by the functionality provided by the PCI DMA interface
+ (see Documentation/PCI/PCI-DMA-mapping.txt). They continue
to be documented below for historical purposes, but new code
must not use them. --davidm 00/12/12 ]
do not have a corresponding kernel virtual address space mapping) and
low-memory pages.
-Note: Please refer to DMA-mapping.txt for a discussion on PCI high mem DMA
-aspects and mapping of scatter gather lists, and support for 64 bit PCI.
+Note: Please refer to Documentation/PCI/PCI-DMA-mapping.txt for a discussion
+on PCI high mem DMA aspects and mapping of scatter gather lists, and support
+for 64 bit PCI.
Special handling is required only for cases where i/o needs to happen on
pages at physical memory addresses beyond what the device can support. In these
API OVERVIEW
The big picture is that USB drivers can continue to ignore most DMA issues,
-though they still must provide DMA-ready buffers (see DMA-mapping.txt).
-That's how they've worked through the 2.4 (and earlier) kernels.
+though they still must provide DMA-ready buffers (see
+Documentation/PCI/PCI-DMA-mapping.txt). That's how they've worked through
+the 2.4 (and earlier) kernels.
OR: they can now be DMA-aware.
force a consistent memory access ordering by using memory barriers. It's
not using a streaming DMA mapping, so it's good for small transfers on
systems where the I/O would otherwise thrash an IOMMU mapping. (See
- Documentation/DMA-mapping.txt for definitions of "coherent" and "streaming"
- DMA mappings.)
+ Documentation/PCI/PCI-DMA-mapping.txt for definitions of "coherent" and
+ "streaming" DMA mappings.)
Asking for 1/Nth of a page (as well as asking for N pages) is reasonably
space-efficient.
Existing buffers aren't usable for DMA without first being mapped into the
DMA address space of the device. However, most buffers passed to your
driver can safely be used with such DMA mapping. (See the first section
-of DMA-mapping.txt, titled "What memory is DMA-able?")
+of Documentation/PCI/PCI-DMA-mapping.txt, titled "What memory is DMA-able?")
- When you're using scatterlists, you can map everything at once. On some
systems, this kicks in an IOMMU and turns the scatterlists into single
* @dir: R/W or both.
* @attrs: optional dma attributes
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
dma_addr_t
sba_map_single_attrs(struct device *dev, void *addr, size_t size, int dir,
* @dir: R/W or both.
* @attrs: optional dma attributes
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
void sba_unmap_single_attrs(struct device *dev, dma_addr_t iova, size_t size,
int dir, struct dma_attrs *attrs)
* @size: number of bytes mapped in driver buffer.
* @dma_handle: IOVA of new buffer.
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
void *
sba_alloc_coherent (struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flags)
* @vaddr: virtual address IOVA of "consistent" buffer.
* @dma_handler: IO virtual address of "consistent" buffer.
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
void sba_free_coherent (struct device *dev, size_t size, void *vaddr, dma_addr_t dma_handle)
{
* @dir: R/W or both.
* @attrs: optional dma attributes
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
int sba_map_sg_attrs(struct device *dev, struct scatterlist *sglist, int nents,
int dir, struct dma_attrs *attrs)
* @dir: R/W or both.
* @attrs: optional dma attributes
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
void sba_unmap_sg_attrs(struct device *dev, struct scatterlist *sglist,
int nents, int dir, struct dma_attrs *attrs)
#include <asm/cacheflush.h>
#include <asm/scatterlist.h>
-/* See Documentation/DMA-mapping.txt */
+/* See Documentation/PCI/PCI-DMA-mapping.txt */
struct hppa_dma_ops {
int (*dma_supported)(struct device *dev, u64 mask);
void *(*alloc_consistent)(struct device *dev, size_t size, dma_addr_t *iova, gfp_t flag);
** PARISC 1.1 Dynamic DMA mapping support.
** This implementation is for PA-RISC platforms that do not support
** I/O TLBs (aka DMA address translation hardware).
-** See Documentation/DMA-mapping.txt for interface definitions.
+** See Documentation/PCI/PCI-DMA-mapping.txt for interface definitions.
**
** (c) Copyright 1999,2000 Hewlett-Packard Company
** (c) Copyright 2000 Grant Grundler
#define _ASM_X86_DMA_MAPPING_H
/*
- * IOMMU interface. See Documentation/DMA-mapping.txt and DMA-API.txt for
- * documentation.
+ * IOMMU interface. See Documentation/PCI/PCI-DMA-mapping.txt and
+ * Documentation/DMA-API.txt for documentation.
*/
#include <linux/scatterlist.h>
* This allows to use PCI devices that only support 32bit addresses on systems
* with more than 4GB.
*
- * See Documentation/DMA-mapping.txt for the interface specification.
+ * See Documentation/PCI/PCI-DMA-mapping.txt for the interface specification.
*
* Copyright 2002 Andi Kleen, SuSE Labs.
* Subject to the GNU General Public License v2 only.
* @dev: instance of PCI owned by the driver that's asking
* @mask: number of address bits this PCI device can handle
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
static int sba_dma_supported( struct device *dev, u64 mask)
{
return(0);
}
- /* Documentation/DMA-mapping.txt tells drivers to try 64-bit first,
- * then fall back to 32-bit if that fails.
+ /* Documentation/PCI/PCI-DMA-mapping.txt tells drivers to try 64-bit
+ * first, then fall back to 32-bit if that fails.
* We are just "encouraging" 32-bit DMA masks here since we can
* never allow IOMMU bypass unless we add special support for ZX1.
*/
* @size: number of bytes to map in driver buffer.
* @direction: R/W or both.
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
static dma_addr_t
sba_map_single(struct device *dev, void *addr, size_t size,
* @size: number of bytes mapped in driver buffer.
* @direction: R/W or both.
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
static void
sba_unmap_single(struct device *dev, dma_addr_t iova, size_t size,
* @size: number of bytes mapped in driver buffer.
* @dma_handle: IOVA of new buffer.
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
static void *sba_alloc_consistent(struct device *hwdev, size_t size,
dma_addr_t *dma_handle, gfp_t gfp)
* @vaddr: virtual address IOVA of "consistent" buffer.
* @dma_handler: IO virtual address of "consistent" buffer.
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
static void
sba_free_consistent(struct device *hwdev, size_t size, void *vaddr,
* @nents: number of entries in list
* @direction: R/W or both.
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
static int
sba_map_sg(struct device *dev, struct scatterlist *sglist, int nents,
* @nents: number of entries in list
* @direction: R/W or both.
*
- * See Documentation/DMA-mapping.txt
+ * See Documentation/PCI/PCI-DMA-mapping.txt
*/
static void
sba_unmap_sg(struct device *dev, struct scatterlist *sglist, int nents,
goto fail;
/* allocate and map coherently-cached memory for a DMA-able buffer */
- /* @see 2.6.26.2/Documentation/DMA-mapping.txt line 318 */
+ /* @see Documentation/PCI/PCI-DMA-mapping.txt, near line 318 */
buffer_virt = (u8 *)pci_alloc_consistent(dev, PAGE_SIZE * 4, &buffer_bus);
if (!buffer_virt) {
printk(KERN_DEBUG "Could not allocate coherent DMA buffer.\n");
#if 1 // @todo For now, disable 64-bit, because I do not understand the implications (DAC!)
/* query for DMA transfer */
- /* @see Documentation/DMA-mapping.txt */
+ /* @see Documentation/PCI/PCI-DMA-mapping.txt */
if (!pci_set_dma_mask(dev, DMA_64BIT_MASK)) {
pci_set_consistent_dma_mask(dev, DMA_64BIT_MASK);
/* use 64-bit DMA */
* does memory allocation too using vmalloc_32().
*
* videobuf_dma_*()
- * see Documentation/DMA-mapping.txt, these functions to
+ * see Documentation/PCI/PCI-DMA-mapping.txt, these functions to
* basically the same. The map function does also build a
* scatterlist for the buffer (and unmap frees it ...)
*
projects / kernel / kernel-generic.git / commitdiff