leandrof/linux
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Documentation: convert PCI-DMA-mapping.txt to use the generic DMA API

Browse Source 
- replace the PCI DMA API (i.e. pci_dma_*) with the generic DMA API.

- make the document more generic (use the PCI specific explanation as
  an example).

[akpm@linux-foundation.org: fix things Randy noticed]
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: "David S. Miller" <davem@davemloft.net>
Reviewed-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
FUJITA Tomonori 2010-03-10 15:23:42 -08:00 committed by Linus Torvalds
parent 5f3cd1e0bb
commit 216bf58f40
1 changed files with 168 additions and 176 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

344
Documentation/PCI/PCI-DMA-mapping.txt
View File

@ -1,12 +1,12 @@
Dynamic DMA mapping Dynamic DMA mapping Guide
=================== =========================
David S. Miller <davem@redhat.com> David S. Miller <davem@redhat.com>
Richard Henderson <rth@cygnus.com> Richard Henderson <rth@cygnus.com>
Jakub Jelinek <jakub@redhat.com> Jakub Jelinek <jakub@redhat.com>
This document describes the DMA mapping system in terms of the pci_ This is a guide to device driver writers on how to use the DMA API
API. For a similar API that works for generic devices, see with example pseudo-code. For a concise description of the API, see
DMA-API.txt. DMA-API.txt.
Most of the 64bit platforms have special hardware that translates bus Most of the 64bit platforms have special hardware that translates bus
@ -26,12 +26,15 @@ mapped only for the time they are actually used and unmapped after the DMA
transfer. transfer.
The following API will work of course even on platforms where no such The following API will work of course even on platforms where no such
hardware exists, see e.g. arch/x86/include/asm/pci.h for how it is implemented on hardware exists.
top of the virt_to_bus interface.
Note that the DMA API works with any bus independent of the underlying
microprocessor architecture. You should use the DMA API rather than
the bus specific DMA API (e.g. pci_dma_*).
First of all, you should make sure First of all, you should make sure
#include <linux/pci.h> #include <linux/dma-mapping.h>
is in your driver. This file will obtain for you the definition of the is in your driver. This file will obtain for you the definition of the
dma_addr_t (which can hold any valid DMA address for the platform) dma_addr_t (which can hold any valid DMA address for the platform)
@ -78,44 +81,43 @@ for you to DMA from/to.
DMA addressing limitations DMA addressing limitations
Does your device have any DMA addressing limitations? For example, is Does your device have any DMA addressing limitations? For example, is
your device only capable of driving the low order 24-bits of address your device only capable of driving the low order 24-bits of address?
on the PCI bus for SAC DMA transfers? If so, you need to inform the If so, you need to inform the kernel of this fact.
PCI layer of this fact.
By default, the kernel assumes that your device can address the full By default, the kernel assumes that your device can address the full
32-bits in a SAC cycle. For a 64-bit DAC capable device, this needs 32-bits. For a 64-bit capable device, this needs to be increased.
to be increased. And for a device with limitations, as discussed in And for a device with limitations, as discussed in the previous
the previous paragraph, it needs to be decreased. paragraph, it needs to be decreased.
pci_alloc_consistent() by default will return 32-bit DMA addresses. Special note about PCI: PCI-X specification requires PCI-X devices to
PCI-X specification requires PCI-X devices to support 64-bit support 64-bit addressing (DAC) for all transactions. And at least
addressing (DAC) for all transactions. And at least one platform (SGI one platform (SGI SN2) requires 64-bit consistent allocations to
SN2) requires 64-bit consistent allocations to operate correctly when operate correctly when the IO bus is in PCI-X mode.
the IO bus is in PCI-X mode. Therefore, like with pci_set_dma_mask(),
it's good practice to call pci_set_consistent_dma_mask() to set the
appropriate mask even if your device only supports 32-bit DMA
(default) and especially if it's a PCI-X device.
For correct operation, you must interrogate the PCI layer in your For correct operation, you must interrogate the kernel in your device
device probe routine to see if the PCI controller on the machine can probe routine to see if the DMA controller on the machine can properly
properly support the DMA addressing limitation your device has. It is support the DMA addressing limitation your device has. It is good
good style to do this even if your device holds the default setting, style to do this even if your device holds the default setting,
because this shows that you did think about these issues wrt. your because this shows that you did think about these issues wrt. your
device. device.
The query is performed via a call to pci_set_dma_mask(): The query is performed via a call to dma_set_mask():
int pci_set_dma_mask(struct pci_dev *pdev, u64 device_mask); int dma_set_mask(struct device *dev, u64 mask);
The query for consistent allocations is performed via a call to The query for consistent allocations is performed via a call to
pci_set_consistent_dma_mask(): dma_set_coherent_mask():
int pci_set_consistent_dma_mask(struct pci_dev *pdev, u64 device_mask); int dma_set_coherent_mask(struct device *dev, u64 mask);
Here, pdev is a pointer to the PCI device struct of your device, and Here, dev is a pointer to the device struct of your device, and mask
device_mask is a bit mask describing which bits of a PCI address your is a bit mask describing which bits of an address your device
device supports. It returns zero if your card can perform DMA supports. It returns zero if your card can perform DMA properly on
properly on the machine given the address mask you provided. the machine given the address mask you provided. In general, the
device struct of your device is embedded in the bus specific device
struct of your device. For example, a pointer to the device struct of
your PCI device is pdev->dev (pdev is a pointer to the PCI device
struct of your device).
If it returns non-zero, your device cannot perform DMA properly on If it returns non-zero, your device cannot perform DMA properly on
this platform, and attempting to do so will result in undefined this platform, and attempting to do so will result in undefined
@ -133,31 +135,30 @@ of your driver reports that performance is bad or that the device is not
even detected, you can ask them for the kernel messages to find out even detected, you can ask them for the kernel messages to find out
exactly why. exactly why.
The standard 32-bit addressing PCI device would do something like The standard 32-bit addressing device would do something like this:
this:
if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) { if (dma_set_mask(dev, DMA_BIT_MASK(32))) {
printk(KERN_WARNING printk(KERN_WARNING
"mydev: No suitable DMA available.\n"); "mydev: No suitable DMA available.\n");
goto ignore_this_device; goto ignore_this_device;
} }
Another common scenario is a 64-bit capable device. The approach Another common scenario is a 64-bit capable device. The approach here
here is to try for 64-bit DAC addressing, but back down to a is to try for 64-bit addressing, but back down to a 32-bit mask that
32-bit mask should that fail. The PCI platform code may fail the should not fail. The kernel may fail the 64-bit mask not because the
64-bit mask not because the platform is not capable of 64-bit platform is not capable of 64-bit addressing. Rather, it may fail in
addressing. Rather, it may fail in this case simply because this case simply because 32-bit addressing is done more efficiently
32-bit SAC addressing is done more efficiently than DAC addressing. than 64-bit addressing. For example, Sparc64 PCI SAC addressing is
Sparc64 is one platform which behaves in this way. more efficient than DAC addressing.
Here is how you would handle a 64-bit capable device which can drive Here is how you would handle a 64-bit capable device which can drive
all 64-bits when accessing streaming DMA: all 64-bits when accessing streaming DMA:
int using_dac; int using_dac;
if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) { if (!dma_set_mask(dev, DMA_BIT_MASK(64))) {
using_dac = 1; using_dac = 1;
} else if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) { } else if (!dma_set_mask(dev, DMA_BIT_MASK(32))) {
using_dac = 0; using_dac = 0;
} else { } else {
printk(KERN_WARNING printk(KERN_WARNING
@ -170,36 +171,36 @@ the case would look like this:
int using_dac, consistent_using_dac; int using_dac, consistent_using_dac;
if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) { if (!dma_set_mask(dev, DMA_BIT_MASK(64))) {
using_dac = 1; using_dac = 1;
consistent_using_dac = 1; consistent_using_dac = 1;
pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64)); dma_set_coherent_mask(dev, DMA_BIT_MASK(64));
} else if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) { } else if (!dma_set_mask(dev, DMA_BIT_MASK(32))) {
using_dac = 0; using_dac = 0;
consistent_using_dac = 0; consistent_using_dac = 0;
pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32)); dma_set_coherent_mask(dev, DMA_BIT_MASK(32));
} else { } else {
printk(KERN_WARNING printk(KERN_WARNING
"mydev: No suitable DMA available.\n"); "mydev: No suitable DMA available.\n");
goto ignore_this_device; goto ignore_this_device;
} }
pci_set_consistent_dma_mask() will always be able to set the same or a dma_set_coherent_mask() will always be able to set the same or a
smaller mask as pci_set_dma_mask(). However for the rare case that a smaller mask as dma_set_mask(). However for the rare case that a
device driver only uses consistent allocations, one would have to device driver only uses consistent allocations, one would have to
check the return value from pci_set_consistent_dma_mask(). check the return value from dma_set_coherent_mask().
Finally, if your device can only drive the low 24-bits of Finally, if your device can only drive the low 24-bits of
address during PCI bus mastering you might do something like: address you might do something like:
if (pci_set_dma_mask(pdev, DMA_BIT_MASK(24))) { if (dma_set_mask(dev, DMA_BIT_MASK(24))) {
printk(KERN_WARNING printk(KERN_WARNING
"mydev: 24-bit DMA addressing not available.\n"); "mydev: 24-bit DMA addressing not available.\n");
goto ignore_this_device; goto ignore_this_device;
} }
When pci_set_dma_mask() is successful, and returns zero, the PCI layer When dma_set_mask() is successful, and returns zero, the kernel saves
saves away this mask you have provided. The PCI layer will use this away this mask you have provided. The kernel will use this
information later when you make DMA mappings. information later when you make DMA mappings.
There is a case which we are aware of at this time, which is worth There is a case which we are aware of at this time, which is worth
@ -208,7 +209,7 @@ functions (for example a sound card provides playback and record
functions) and the various different functions have _different_ functions) and the various different functions have _different_
DMA addressing limitations, you may wish to probe each mask and DMA addressing limitations, you may wish to probe each mask and
only provide the functionality which the machine can handle. It only provide the functionality which the machine can handle. It
is important that the last call to pci_set_dma_mask() be for the is important that the last call to dma_set_mask() be for the
most specific mask. most specific mask.
Here is pseudo-code showing how this might be done: Here is pseudo-code showing how this might be done:
@ -217,17 +218,17 @@ Here is pseudo-code showing how this might be done:
#define RECORD_ADDRESS_BITS DMA_BIT_MASK(24) #define RECORD_ADDRESS_BITS DMA_BIT_MASK(24)
struct my_sound_card *card; struct my_sound_card *card;
struct pci_dev *pdev; struct device *dev;
... ...
if (!pci_set_dma_mask(pdev, PLAYBACK_ADDRESS_BITS)) { if (!dma_set_mask(dev, PLAYBACK_ADDRESS_BITS)) {
card->playback_enabled = 1; card->playback_enabled = 1;
} else { } else {
card->playback_enabled = 0; card->playback_enabled = 0;
printk(KERN_WARNING "%s: Playback disabled due to DMA limitations.\n", printk(KERN_WARNING "%s: Playback disabled due to DMA limitations.\n",
card->name); card->name);
} }
if (!pci_set_dma_mask(pdev, RECORD_ADDRESS_BITS)) { if (!dma_set_mask(dev, RECORD_ADDRESS_BITS)) {
card->record_enabled = 1; card->record_enabled = 1;
} else { } else {
card->record_enabled = 0; card->record_enabled = 0;
@ -252,8 +253,8 @@ There are two types of DMA mappings:
Think of "consistent" as "synchronous" or "coherent". Think of "consistent" as "synchronous" or "coherent".
The current default is to return consistent memory in the low 32 The current default is to return consistent memory in the low 32
bits of the PCI bus space. However, for future compatibility you bits of the bus space. However, for future compatibility you should
should set the consistent mask even if this default is fine for your set the consistent mask even if this default is fine for your
driver. driver.
Good examples of what to use consistent mappings for are: Good examples of what to use consistent mappings for are:
@ -285,9 +286,9 @@ There are two types of DMA mappings:
found in PCI bridges (such as by reading a register's value found in PCI bridges (such as by reading a register's value
after writing it). after writing it).
- Streaming DMA mappings which are usually mapped for one DMA transfer, - Streaming DMA mappings which are usually mapped for one DMA
unmapped right after it (unless you use pci_dma_sync_* below) and for which transfer, unmapped right after it (unless you use dma_sync_* below)
hardware can optimize for sequential accesses. and for which hardware can optimize for sequential accesses.
This of "streaming" as "asynchronous" or "outside the coherency This of "streaming" as "asynchronous" or "outside the coherency
domain". domain".
@ -302,8 +303,8 @@ There are two types of DMA mappings:
optimizations the hardware allows. To this end, when using optimizations the hardware allows. To this end, when using
such mappings you must be explicit about what you want to happen. such mappings you must be explicit about what you want to happen.
Neither type of DMA mapping has alignment restrictions that come Neither type of DMA mapping has alignment restrictions that come from
from PCI, although some devices may have such restrictions. the underlying bus, although some devices may have such restrictions.
Also, systems with caches that aren't DMA-coherent will work better Also, systems with caches that aren't DMA-coherent will work better
when the underlying buffers don't share cache lines with other data. when the underlying buffers don't share cache lines with other data.
@ -315,33 +316,27 @@ you should do:
dma_addr_t dma_handle; dma_addr_t dma_handle;
cpu_addr = pci_alloc_consistent(pdev, size, &dma_handle); cpu_addr = dma_alloc_coherent(dev, size, &dma_handle, gfp);
where pdev is a struct pci_dev *. This may be called in interrupt context. where device is a struct device *. This may be called in interrupt
You should use dma_alloc_coherent (see DMA-API.txt) for buses context with the GFP_ATOMIC flag.
where devices don't have struct pci_dev (like ISA, EISA).
This argument is needed because the DMA translations may be bus
specific (and often is private to the bus which the device is attached
to).
Size is the length of the region you want to allocate, in bytes. Size is the length of the region you want to allocate, in bytes.
This routine will allocate RAM for that region, so it acts similarly to This routine will allocate RAM for that region, so it acts similarly to
__get_free_pages (but takes size instead of a page order). If your __get_free_pages (but takes size instead of a page order). If your
driver needs regions sized smaller than a page, you may prefer using driver needs regions sized smaller than a page, you may prefer using
the pci_pool interface, described below. the dma_pool interface, described below.
The consistent DMA mapping interfaces, for non-NULL pdev, will by The consistent DMA mapping interfaces, for non-NULL dev, will by
default return a DMA address which is SAC (Single Address Cycle) default return a DMA address which is 32-bit addressable. Even if the
addressable. Even if the device indicates (via PCI dma mask) that it device indicates (via DMA mask) that it may address the upper 32-bits,
may address the upper 32-bits and thus perform DAC cycles, consistent consistent allocation will only return > 32-bit addresses for DMA if
allocation will only return > 32-bit PCI addresses for DMA if the the consistent DMA mask has been explicitly changed via
consistent dma mask has been explicitly changed via dma_set_coherent_mask(). This is true of the dma_pool interface as
pci_set_consistent_dma_mask(). This is true of the pci_pool interface well.
as well.
pci_alloc_consistent returns two values: the virtual address which you dma_alloc_coherent returns two values: the virtual address which you
can use to access it from the CPU and dma_handle which you pass to the can use to access it from the CPU and dma_handle which you pass to the
card. card.
@ -354,54 +349,54 @@ buffer you receive will not cross a 64K boundary.
To unmap and free such a DMA region, you call: To unmap and free such a DMA region, you call:
pci_free_consistent(pdev, size, cpu_addr, dma_handle); dma_free_coherent(dev, size, cpu_addr, dma_handle);
where pdev, size are the same as in the above call and cpu_addr and where dev, size are the same as in the above call and cpu_addr and
dma_handle are the values pci_alloc_consistent returned to you. dma_handle are the values dma_alloc_coherent returned to you.
This function may not be called in interrupt context. This function may not be called in interrupt context.
If your driver needs lots of smaller memory regions, you can write If your driver needs lots of smaller memory regions, you can write
custom code to subdivide pages returned by pci_alloc_consistent, custom code to subdivide pages returned by dma_alloc_coherent,
or you can use the pci_pool API to do that. A pci_pool is like or you can use the dma_pool API to do that. A dma_pool is like
a kmem_cache, but it uses pci_alloc_consistent not __get_free_pages. a kmem_cache, but it uses dma_alloc_coherent not __get_free_pages.
Also, it understands common hardware constraints for alignment, Also, it understands common hardware constraints for alignment,
like queue heads needing to be aligned on N byte boundaries. like queue heads needing to be aligned on N byte boundaries.
Create a pci_pool like this: Create a dma_pool like this:
struct pci_pool *pool; struct dma_pool *pool;
pool = pci_pool_create(name, pdev, size, align, alloc); pool = dma_pool_create(name, dev, size, align, alloc);
The "name" is for diagnostics (like a kmem_cache name); pdev and size The "name" is for diagnostics (like a kmem_cache name); dev and size
are as above. The device's hardware alignment requirement for this are as above. The device's hardware alignment requirement for this
type of data is "align" (which is expressed in bytes, and must be a type of data is "align" (which is expressed in bytes, and must be a
power of two). If your device has no boundary crossing restrictions, power of two). If your device has no boundary crossing restrictions,
pass 0 for alloc; passing 4096 says memory allocated from this pool pass 0 for alloc; passing 4096 says memory allocated from this pool
must not cross 4KByte boundaries (but at that time it may be better to must not cross 4KByte boundaries (but at that time it may be better to
go for pci_alloc_consistent directly instead). go for dma_alloc_coherent directly instead).
Allocate memory from a pci pool like this: Allocate memory from a dma pool like this:
cpu_addr = pci_pool_alloc(pool, flags, &dma_handle); cpu_addr = dma_pool_alloc(pool, flags, &dma_handle);
flags are SLAB_KERNEL if blocking is permitted (not in_interrupt nor flags are SLAB_KERNEL if blocking is permitted (not in_interrupt nor
holding SMP locks), SLAB_ATOMIC otherwise. Like pci_alloc_consistent, holding SMP locks), SLAB_ATOMIC otherwise. Like dma_alloc_coherent,
this returns two values, cpu_addr and dma_handle. this returns two values, cpu_addr and dma_handle.
Free memory that was allocated from a pci_pool like this: Free memory that was allocated from a dma_pool like this:
pci_pool_free(pool, cpu_addr, dma_handle); dma_pool_free(pool, cpu_addr, dma_handle);
where pool is what you passed to pci_pool_alloc, and cpu_addr and where pool is what you passed to dma_pool_alloc, and cpu_addr and
dma_handle are the values pci_pool_alloc returned. This function dma_handle are the values dma_pool_alloc returned. This function
may be called in interrupt context. may be called in interrupt context.
Destroy a pci_pool by calling: Destroy a dma_pool by calling:
pci_pool_destroy(pool); dma_pool_destroy(pool);
Make sure you've called pci_pool_free for all memory allocated Make sure you've called dma_pool_free for all memory allocated
from a pool before you destroy the pool. This function may not from a pool before you destroy the pool. This function may not
be called in interrupt context. be called in interrupt context.
@ -411,15 +406,15 @@ The interfaces described in subsequent portions of this document
take a DMA direction argument, which is an integer and takes on take a DMA direction argument, which is an integer and takes on
one of the following values: one of the following values:
PCI_DMA_BIDIRECTIONAL DMA_BIDIRECTIONAL
PCI_DMA_TODEVICE DMA_TO_DEVICE
PCI_DMA_FROMDEVICE DMA_FROM_DEVICE
PCI_DMA_NONE DMA_NONE
One should provide the exact DMA direction if you know it. One should provide the exact DMA direction if you know it.
PCI_DMA_TODEVICE means "from main memory to the PCI device" DMA_TO_DEVICE means "from main memory to the device"
PCI_DMA_FROMDEVICE means "from the PCI device to main memory" DMA_FROM_DEVICE means "from the device to main memory"
It is the direction in which the data moves during the DMA It is the direction in which the data moves during the DMA
transfer. transfer.
@ -427,12 +422,12 @@ You are _strongly_ encouraged to specify this as precisely
as you possibly can. as you possibly can.
If you absolutely cannot know the direction of the DMA transfer, If you absolutely cannot know the direction of the DMA transfer,
specify PCI_DMA_BIDIRECTIONAL. It means that the DMA can go in specify DMA_BIDIRECTIONAL. It means that the DMA can go in
either direction. The platform guarantees that you may legally either direction. The platform guarantees that you may legally
specify this, and that it will work, but this may be at the specify this, and that it will work, but this may be at the
cost of performance for example. cost of performance for example.
The value PCI_DMA_NONE is to be used for debugging. One can The value DMA_NONE is to be used for debugging. One can
hold this in a data structure before you come to know the hold this in a data structure before you come to know the
precise direction, and this will help catch cases where your precise direction, and this will help catch cases where your
direction tracking logic has failed to set things up properly. direction tracking logic has failed to set things up properly.
@ -442,21 +437,21 @@ potential platform-specific optimizations of such) is for debugging.
Some platforms actually have a write permission boolean which DMA Some platforms actually have a write permission boolean which DMA
mappings can be marked with, much like page protections in the user mappings can be marked with, much like page protections in the user
program address space. Such platforms can and do report errors in the program address space. Such platforms can and do report errors in the
kernel logs when the PCI controller hardware detects violation of the kernel logs when the DMA controller hardware detects violation of the
permission setting. permission setting.
Only streaming mappings specify a direction, consistent mappings Only streaming mappings specify a direction, consistent mappings
implicitly have a direction attribute setting of implicitly have a direction attribute setting of
PCI_DMA_BIDIRECTIONAL. DMA_BIDIRECTIONAL.
The SCSI subsystem tells you the direction to use in the The SCSI subsystem tells you the direction to use in the
'sc_data_direction' member of the SCSI command your driver is 'sc_data_direction' member of the SCSI command your driver is
working on. working on.
For Networking drivers, it's a rather simple affair. For transmit For Networking drivers, it's a rather simple affair. For transmit
packets, map/unmap them with the PCI_DMA_TODEVICE direction packets, map/unmap them with the DMA_TO_DEVICE direction
specifier. For receive packets, just the opposite, map/unmap them specifier. For receive packets, just the opposite, map/unmap them
with the PCI_DMA_FROMDEVICE direction specifier. with the DMA_FROM_DEVICE direction specifier.
Using Streaming DMA mappings Using Streaming DMA mappings
@ -467,43 +462,43 @@ scatterlist.
To map a single region, you do: To map a single region, you do:
struct pci_dev *pdev = mydev->pdev; struct device *dev = &my_dev->dev;
dma_addr_t dma_handle; dma_addr_t dma_handle;
void *addr = buffer->ptr; void *addr = buffer->ptr;
size_t size = buffer->len; size_t size = buffer->len;
dma_handle = pci_map_single(pdev, addr, size, direction); dma_handle = dma_map_single(dev, addr, size, direction);
and to unmap it: and to unmap it:
pci_unmap_single(pdev, dma_handle, size, direction); dma_unmap_single(dev, dma_handle, size, direction);
You should call pci_unmap_single when the DMA activity is finished, e.g. You should call dma_unmap_single when the DMA activity is finished, e.g.
from the interrupt which told you that the DMA transfer is done. from the interrupt which told you that the DMA transfer is done.
Using cpu pointers like this for single mappings has a disadvantage, Using cpu pointers like this for single mappings has a disadvantage,
you cannot reference HIGHMEM memory in this way. Thus, there is a you cannot reference HIGHMEM memory in this way. Thus, there is a
map/unmap interface pair akin to pci_{map,unmap}_single. These map/unmap interface pair akin to dma_{map,unmap}_single. These
interfaces deal with page/offset pairs instead of cpu pointers. interfaces deal with page/offset pairs instead of cpu pointers.
Specifically: Specifically:
struct pci_dev *pdev = mydev->pdev; struct device *dev = &my_dev->dev;
dma_addr_t dma_handle; dma_addr_t dma_handle;
struct page *page = buffer->page; struct page *page = buffer->page;
unsigned long offset = buffer->offset; unsigned long offset = buffer->offset;
size_t size = buffer->len; size_t size = buffer->len;
dma_handle = pci_map_page(pdev, page, offset, size, direction); dma_handle = dma_map_page(dev, page, offset, size, direction);
... ...
pci_unmap_page(pdev, dma_handle, size, direction); dma_unmap_page(dev, dma_handle, size, direction);
Here, "offset" means byte offset within the given page. Here, "offset" means byte offset within the given page.
With scatterlists, you map a region gathered from several regions by: With scatterlists, you map a region gathered from several regions by:
int i, count = pci_map_sg(pdev, sglist, nents, direction); int i, count = dma_map_sg(dev, sglist, nents, direction);
struct scatterlist *sg; struct scatterlist *sg;
for_each_sg(sglist, sg, count, i) { for_each_sg(sglist, sg, count, i) {
@ -527,16 +522,16 @@ accessed sg->address and sg->length as shown above.
To unmap a scatterlist, just call: To unmap a scatterlist, just call:
pci_unmap_sg(pdev, sglist, nents, direction); dma_unmap_sg(dev, sglist, nents, direction);
Again, make sure DMA activity has already finished. Again, make sure DMA activity has already finished.
PLEASE NOTE: The 'nents' argument to the pci_unmap_sg call must be PLEASE NOTE: The 'nents' argument to the dma_unmap_sg call must be
the _same_ one you passed into the pci_map_sg call, the _same_ one you passed into the dma_map_sg call,
it should _NOT_ be the 'count' value _returned_ from the it should _NOT_ be the 'count' value _returned_ from the
pci_map_sg call. dma_map_sg call.
Every pci_map_{single,sg} call should have its pci_unmap_{single,sg} Every dma_map_{single,sg} call should have its dma_unmap_{single,sg}
counterpart, because the bus address space is a shared resource (although counterpart, because the bus address space is a shared resource (although
in some ports the mapping is per each BUS so less devices contend for the in some ports the mapping is per each BUS so less devices contend for the
same bus address space) and you could render the machine unusable by eating same bus address space) and you could render the machine unusable by eating
@ -547,14 +542,14 @@ the data in between the DMA transfers, the buffer needs to be synced
properly in order for the cpu and device to see the most uptodate and properly in order for the cpu and device to see the most uptodate and
correct copy of the DMA buffer. correct copy of the DMA buffer.
So, firstly, just map it with pci_map_{single,sg}, and after each DMA So, firstly, just map it with dma_map_{single,sg}, and after each DMA
transfer call either: transfer call either:
pci_dma_sync_single_for_cpu(pdev, dma_handle, size, direction); dma_sync_single_for_cpu(dev, dma_handle, size, direction);
or: or:
pci_dma_sync_sg_for_cpu(pdev, sglist, nents, direction); dma_sync_sg_for_cpu(dev, sglist, nents, direction);
as appropriate. as appropriate.
@ -562,27 +557,27 @@ Then, if you wish to let the device get at the DMA area again,
finish accessing the data with the cpu, and then before actually finish accessing the data with the cpu, and then before actually
giving the buffer to the hardware call either: giving the buffer to the hardware call either:
pci_dma_sync_single_for_device(pdev, dma_handle, size, direction); dma_sync_single_for_device(dev, dma_handle, size, direction);
or: or:
pci_dma_sync_sg_for_device(dev, sglist, nents, direction); dma_sync_sg_for_device(dev, sglist, nents, direction);
as appropriate. as appropriate.
After the last DMA transfer call one of the DMA unmap routines After the last DMA transfer call one of the DMA unmap routines
pci_unmap_{single,sg}. If you don't touch the data from the first pci_map_* dma_unmap_{single,sg}. If you don't touch the data from the first dma_map_*
call till pci_unmap_*, then you don't have to call the pci_dma_sync_* call till dma_unmap_*, then you don't have to call the dma_sync_*
routines at all. routines at all.
Here is pseudo code which shows a situation in which you would need Here is pseudo code which shows a situation in which you would need
to use the pci_dma_sync_*() interfaces. to use the dma_sync_*() interfaces.
my_card_setup_receive_buffer(struct my_card *cp, char *buffer, int len) my_card_setup_receive_buffer(struct my_card *cp, char *buffer, int len)
{ {
dma_addr_t mapping; dma_addr_t mapping;
mapping = pci_map_single(cp->pdev, buffer, len, PCI_DMA_FROMDEVICE); mapping = dma_map_single(cp->dev, buffer, len, DMA_FROM_DEVICE);
cp->rx_buf = buffer; cp->rx_buf = buffer;
cp->rx_len = len; cp->rx_len = len;
@ -606,25 +601,25 @@ to use the pci_dma_sync_*() interfaces.
* the DMA transfer with the CPU first * the DMA transfer with the CPU first
* so that we see updated contents. * so that we see updated contents.
*/ */
pci_dma_sync_single_for_cpu(cp->pdev, cp->rx_dma, dma_sync_single_for_cpu(&cp->dev, cp->rx_dma,
cp->rx_len, cp->rx_len,
PCI_DMA_FROMDEVICE); DMA_FROM_DEVICE);
/* Now it is safe to examine the buffer. */ /* Now it is safe to examine the buffer. */
hp = (struct my_card_header *) cp->rx_buf; hp = (struct my_card_header *) cp->rx_buf;
if (header_is_ok(hp)) { if (header_is_ok(hp)) {
pci_unmap_single(cp->pdev, cp->rx_dma, cp->rx_len, dma_unmap_single(&cp->dev, cp->rx_dma, cp->rx_len,
PCI_DMA_FROMDEVICE); DMA_FROM_DEVICE);
pass_to_upper_layers(cp->rx_buf); pass_to_upper_layers(cp->rx_buf);
make_and_setup_new_rx_buf(cp); make_and_setup_new_rx_buf(cp);
} else { } else {
/* Just sync the buffer and give it back /* Just sync the buffer and give it back
* to the card. * to the card.
*/ */
pci_dma_sync_single_for_device(cp->pdev, dma_sync_single_for_device(&cp->dev,
cp->rx_dma, cp->rx_dma,
cp->rx_len, cp->rx_len,
PCI_DMA_FROMDEVICE); DMA_FROM_DEVICE);
give_rx_buf_to_card(cp); give_rx_buf_to_card(cp);
} }
} }
@ -634,19 +629,19 @@ Drivers converted fully to this interface should not use virt_to_bus any
longer, nor should they use bus_to_virt. Some drivers have to be changed a longer, nor should they use bus_to_virt. Some drivers have to be changed a
little bit, because there is no longer an equivalent to bus_to_virt in the little bit, because there is no longer an equivalent to bus_to_virt in the
dynamic DMA mapping scheme - you have to always store the DMA addresses dynamic DMA mapping scheme - you have to always store the DMA addresses
returned by the pci_alloc_consistent, pci_pool_alloc, and pci_map_single returned by the dma_alloc_coherent, dma_pool_alloc, and dma_map_single
calls (pci_map_sg stores them in the scatterlist itself if the platform calls (dma_map_sg stores them in the scatterlist itself if the platform
supports dynamic DMA mapping in hardware) in your driver structures and/or supports dynamic DMA mapping in hardware) in your driver structures and/or
in the card registers. in the card registers.
All PCI drivers should be using these interfaces with no exceptions. All drivers should be using these interfaces with no exceptions. It
It is planned to completely remove virt_to_bus() and bus_to_virt() as is planned to completely remove virt_to_bus() and bus_to_virt() as
they are entirely deprecated. Some ports already do not provide these they are entirely deprecated. Some ports already do not provide these
as it is impossible to correctly support them. as it is impossible to correctly support them.
Optimizing Unmap State Space Consumption Optimizing Unmap State Space Consumption
On many platforms, pci_unmap_{single,page}() is simply a nop. On many platforms, dma_unmap_{single,page}() is simply a nop.
Therefore, keeping track of the mapping address and length is a waste Therefore, keeping track of the mapping address and length is a waste
of space. Instead of filling your drivers up with ifdefs and the like of space. Instead of filling your drivers up with ifdefs and the like
to "work around" this (which would defeat the whole purpose of a to "work around" this (which would defeat the whole purpose of a
@ -655,7 +650,7 @@ portable API) the following facilities are provided.
Actually, instead of describing the macros one by one, we'll Actually, instead of describing the macros one by one, we'll
transform some example code. transform some example code.
1) Use DECLARE_PCI_UNMAP_{ADDR,LEN} in state saving structures. 1) Use DEFINE_DMA_UNMAP_{ADDR,LEN} in state saving structures.
Example, before: Example, before:
struct ring_state { struct ring_state {
@ -668,14 +663,11 @@ transform some example code.
struct ring_state { struct ring_state {
struct sk_buff *skb; struct sk_buff *skb;
DECLARE_PCI_UNMAP_ADDR(mapping) DEFINE_DMA_UNMAP_ADDR(mapping);
DECLARE_PCI_UNMAP_LEN(len) DEFINE_DMA_UNMAP_LEN(len);
}; };
NOTE: DO NOT put a semicolon at the end of the DECLARE_*() 2) Use dma_unmap_{addr,len}_set to set these values.
macro.
2) Use pci_unmap_{addr,len}_set to set these values.
Example, before: Example, before:
ringp->mapping = FOO; ringp->mapping = FOO;
@ -683,21 +675,21 @@ transform some example code.
after: after:
pci_unmap_addr_set(ringp, mapping, FOO); dma_unmap_addr_set(ringp, mapping, FOO);
pci_unmap_len_set(ringp, len, BAR); dma_unmap_len_set(ringp, len, BAR);
3) Use pci_unmap_{addr,len} to access these values. 3) Use dma_unmap_{addr,len} to access these values.
Example, before: Example, before:
pci_unmap_single(pdev, ringp->mapping, ringp->len, dma_unmap_single(dev, ringp->mapping, ringp->len,
PCI_DMA_FROMDEVICE); DMA_FROM_DEVICE);
after: after:
pci_unmap_single(pdev, dma_unmap_single(dev,
pci_unmap_addr(ringp, mapping), dma_unmap_addr(ringp, mapping),
pci_unmap_len(ringp, len), dma_unmap_len(ringp, len),
PCI_DMA_FROMDEVICE); DMA_FROM_DEVICE);
It really should be self-explanatory. We treat the ADDR and LEN It really should be self-explanatory. We treat the ADDR and LEN
separately, because it is possible for an implementation to only separately, because it is possible for an implementation to only
@ -732,15 +724,15 @@ to "Closing".
DMA address space is limited on some architectures and an allocation DMA address space is limited on some architectures and an allocation
failure can be determined by: failure can be determined by:
- checking if pci_alloc_consistent returns NULL or pci_map_sg returns 0 - checking if dma_alloc_coherent returns NULL or dma_map_sg returns 0
- checking the returned dma_addr_t of pci_map_single and pci_map_page - checking the returned dma_addr_t of dma_map_single and dma_map_page
by using pci_dma_mapping_error(): by using dma_mapping_error():
dma_addr_t dma_handle; dma_addr_t dma_handle;
dma_handle = pci_map_single(pdev, addr, size, direction); dma_handle = dma_map_single(dev, addr, size, direction);
if (pci_dma_mapping_error(pdev, dma_handle)) { if (dma_mapping_error(dev, dma_handle)) {
/* /*
* reduce current DMA mapping usage, * reduce current DMA mapping usage,
* delay and try again later or * delay and try again later or