diff options
| author | Ingo Molnar <mingo@elte.hu> | 2009-01-31 11:32:31 -0500 |
|---|---|---|
| committer | Ingo Molnar <mingo@elte.hu> | 2009-02-05 16:30:01 -0500 |
| commit | 9d45cf9e36bf9bcf16df6e1cbf049807c8402823 (patch) | |
| tree | 2118a16701418af10d215d2174df7ee0a5cbe6bd /Documentation/usb/dma.txt | |
| parent | a146649bc19d5eba4f5bfac6720c5f252d517a71 (diff) | |
| parent | 0cd5c3c80a0ebd68c08312fa7d8c13149cc61c4c (diff) | |
Merge branch 'x86/urgent' into x86/apic
Conflicts:
arch/x86/mach-default/setup.c
Semantic merge:
arch/x86/kernel/irqinit_32.c
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Diffstat (limited to 'Documentation/usb/dma.txt')
| -rw-r--r-- | Documentation/usb/dma.txt | 11 |
1 files changed, 6 insertions, 5 deletions
diff --git a/Documentation/usb/dma.txt b/Documentation/usb/dma.txt index e8b50b7de9d9..cfdcd16e3abf 100644 --- a/Documentation/usb/dma.txt +++ b/Documentation/usb/dma.txt | |||
| @@ -6,8 +6,9 @@ in the kernel usb programming guide (kerneldoc, from the source code). | |||
| 6 | API OVERVIEW | 6 | API OVERVIEW |
| 7 | 7 | ||
| 8 | The big picture is that USB drivers can continue to ignore most DMA issues, | 8 | The big picture is that USB drivers can continue to ignore most DMA issues, |
| 9 | though they still must provide DMA-ready buffers (see DMA-mapping.txt). | 9 | though they still must provide DMA-ready buffers (see |
| 10 | That's how they've worked through the 2.4 (and earlier) kernels. | 10 | Documentation/PCI/PCI-DMA-mapping.txt). That's how they've worked through |
| 11 | the 2.4 (and earlier) kernels. | ||
| 11 | 12 | ||
| 12 | OR: they can now be DMA-aware. | 13 | OR: they can now be DMA-aware. |
| 13 | 14 | ||
| @@ -62,8 +63,8 @@ and effects like cache-trashing can impose subtle penalties. | |||
| 62 | force a consistent memory access ordering by using memory barriers. It's | 63 | force a consistent memory access ordering by using memory barriers. It's |
| 63 | not using a streaming DMA mapping, so it's good for small transfers on | 64 | not using a streaming DMA mapping, so it's good for small transfers on |
| 64 | systems where the I/O would otherwise thrash an IOMMU mapping. (See | 65 | systems where the I/O would otherwise thrash an IOMMU mapping. (See |
| 65 | Documentation/DMA-mapping.txt for definitions of "coherent" and "streaming" | 66 | Documentation/PCI/PCI-DMA-mapping.txt for definitions of "coherent" and |
| 66 | DMA mappings.) | 67 | "streaming" DMA mappings.) |
| 67 | 68 | ||
| 68 | Asking for 1/Nth of a page (as well as asking for N pages) is reasonably | 69 | Asking for 1/Nth of a page (as well as asking for N pages) is reasonably |
| 69 | space-efficient. | 70 | space-efficient. |
| @@ -93,7 +94,7 @@ WORKING WITH EXISTING BUFFERS | |||
| 93 | Existing buffers aren't usable for DMA without first being mapped into the | 94 | Existing buffers aren't usable for DMA without first being mapped into the |
| 94 | DMA address space of the device. However, most buffers passed to your | 95 | DMA address space of the device. However, most buffers passed to your |
| 95 | driver can safely be used with such DMA mapping. (See the first section | 96 | driver can safely be used with such DMA mapping. (See the first section |
| 96 | of DMA-mapping.txt, titled "What memory is DMA-able?") | 97 | of Documentation/PCI/PCI-DMA-mapping.txt, titled "What memory is DMA-able?") |
| 97 | 98 | ||
| 98 | - When you're using scatterlists, you can map everything at once. On some | 99 | - When you're using scatterlists, you can map everything at once. On some |
| 99 | systems, this kicks in an IOMMU and turns the scatterlists into single | 100 | systems, this kicks in an IOMMU and turns the scatterlists into single |