linux-toradex.git/Documentation/core-api/dma-api.rst, branch v6.1

dma-mapping: add dma_opt_mapping_size()

2022-07-19T04:05:41+00:00

Streaming DMA mapping involving an IOMMU may be much slower for larger
total mapping size. This is because every IOMMU DMA mapping requires an
IOVA to be allocated and freed. IOVA sizes above a certain limit are not
cached, which can have a big impact on DMA mapping performance.

Provide an API for device drivers to know this "optimal" limit, such that
they may try to produce mapping which don't exceed it.

Signed-off-by: John Garry 
Reviewed-by: Damien Le Moal 
Acked-by: Martin K. Petersen 
Signed-off-by: Christoph Hellwig

docs: core-api: avoid using ReST :doc:`foo` markup

2021-06-17T19:24:37+00:00

The :doc:`foo` tag is auto-generated via automarkup.py.
So, use the filename at the sources, instead of :doc:`foo`.

Signed-off-by: Mauro Carvalho Chehab 
Link: https://lore.kernel.org/r/d967d490b6655735b7df292f88859b5a1b07d0d7.1623824363.git.mchehab+huawei@kernel.org
Signed-off-by: Jonathan Corbet

dma-mapping: add a dma_alloc_noncontiguous API

2021-03-15T09:02:31+00:00

Add a new API that returns a potentiall virtually non-contigous sg_table
and a DMA address.  This API is only properly implemented for dma-iommu
and will simply return a contigious chunk as a fallback.

The intent is that drivers can use this API if either:

 - no kernel mapping or only temporary kernel mappings are required.
   That is as a better replacement for DMA_ATTR_NO_KERNEL_MAPPING
 - a kernel mapping is required for cached and DMA mapped pages, but
   the driver also needs the pages to e.g. map them to userspace.
   In that sense it is a replacement for some aspects of the recently
   removed and never fully implemented DMA_ATTR_NON_CONSISTENT

Signed-off-by: Christoph Hellwig 
Reviewed-by: Tomasz Figa 
Tested-by: Ricardo Ribalda

dma-mapping: add a dma_mmap_pages helper

2021-03-15T09:02:31+00:00

Add a helper to map memory allocated using dma_alloc_pages into
a user address space, similar to the dma_alloc_attrs function for
coherent allocations.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Tomasz Figa 
Tested-by: Ricardo Ribalda

dma-mapping: remove the {alloc,free}_noncoherent methods

2021-02-09T17:01:38+00:00

It turns out allowing non-contigous allocations here was a rather bad
idea, as we'll now need to define ways to get the pages for mmaping
or dma_buf sharing.  Revert this change and stick to the original
concept.  A different API for the use case of non-contigous allocations
will be added back later.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Tomasz Figa 
Tested-by: Ricardo Ribalda :wq

dma-mapping: document dma_{alloc,free}_pages

2020-10-23T10:07:46+00:00

Document the new dma_alloc_pages and dma_free_pages APIs, and fix
up the documentation for dma_alloc_noncoherent and dma_free_noncoherent.

Reported-by: Robin Murphy 
Signed-off-by: Christoph Hellwig 
Reviewed-by: Robin Murphy

dma-mapping: add a new dma_alloc_noncoherent API

2020-09-25T04:20:43+00:00

Add a new API to allocate and free memory that is guaranteed to be
addressable by a device, but which potentially is not cache coherent
for DMA.

To transfer ownership to and from the device, the existing streaming
DMA API calls dma_sync_single_for_device and dma_sync_single_for_cpu
must be used.

For now the new calls are implemented on top of dma_alloc_attrs just
like the old-noncoherent API, but once all drivers are switched to
the new API it will be replaced with a better working implementation
that is available on all architectures.

Signed-off-by: Christoph Hellwig

dma-mapping: move the dma_declare_coherent_memory documentation

2020-09-11T07:17:42+00:00

dma_declare_coherent_memory should not be in a DMA API guide aimed
at driver writers (that is consumers of the API).  Move it to a comment
near the function instead.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Robin Murphy

Merge tag 'docs-5.9' of git://git.lwn.net/linux

2020-08-05T05:47:54+00:00

Pull documentation updates from Jonathan Corbet:
 "It's been a busy cycle for documentation - hopefully the busiest for a
  while to come. Changes include:

   - Some new Chinese translations

   - Progress on the battle against double words words and non-HTTPS
     URLs

   - Some block-mq documentation

   - More RST conversions from Mauro. At this point, that task is
     essentially complete, so we shouldn't see this kind of churn again
     for a while. Unless we decide to switch to asciidoc or
     something...:)

   - Lots of typo fixes, warning fixes, and more"

* tag 'docs-5.9' of git://git.lwn.net/linux: (195 commits)
  scripts/kernel-doc: optionally treat warnings as errors
  docs: ia64: correct typo
  mailmap: add entry for 
  doc/zh_CN: add cpu-load Chinese version
  Documentation/admin-guide: tainted-kernels: fix spelling mistake
  MAINTAINERS: adjust kprobes.rst entry to new location
  devices.txt: document rfkill allocation
  PCI: correct flag name
  docs: filesystems: vfs: correct flag name
  docs: filesystems: vfs: correct sync_mode flag names
  docs: path-lookup: markup fixes for emphasis
  docs: path-lookup: more markup fixes
  docs: path-lookup: fix HTML entity mojibake
  CREDITS: Replace HTTP links with HTTPS ones
  docs: process: Add an example for creating a fixes tag
  doc/zh_CN: add Chinese translation prefer section
  doc/zh_CN: add clearing-warn-once Chinese version
  doc/zh_CN: add admin-guide index
  doc:it_IT: process: coding-style.rst: Correct __maybe_unused compiler label
  futex: MAINTAINERS: Re-add selftests directory
  ...

dma-mapping: Add a new dma_need_sync API

2020-06-30T13:44:03+00:00

Add a new API to check if calls to dma_sync_single_for_{device,cpu} are
required for a given DMA streaming mapping.

Signed-off-by: Christoph Hellwig 
Signed-off-by: Daniel Borkmann 
Link: https://lore.kernel.org/bpf/20200629130359.2690853-2-hch@lst.de