Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1690030 > unrolled thread

[PATCH v3 2/4] i2c: add docs to clarify DMA handling

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [PATCH v3 2/4] i2c: add docs to clarify DMA handling Wolfram Sang <wsa+renesas@sang-engineering.com> - 2017-07-18 12:30 +0200
    Re: [PATCH v3 2/4] i2c: add docs to clarify DMA handling Niklas Söderlund          <niklas.soderlund@ragnatech.se> - 2017-07-19 11:30 +0200
    Re: [PATCH v3 2/4] i2c: add docs to clarify DMA handling Jonathan Cameron <jic23@jic23.retrosnub.co.uk> - 2017-07-23 13:30 +0200

#1690030 — [PATCH v3 2/4] i2c: add docs to clarify DMA handling

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
---
Changes since v2:

* documentation updates. Hopefully better wording now

 Documentation/i2c/DMA-considerations | 38 ++++++++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
 create mode 100644 Documentation/i2c/DMA-considerations

diff --git a/Documentation/i2c/DMA-considerations b/Documentation/i2c/DMA-considerations
new file mode 100644
index 00000000000000..e46c24d65c8556
--- /dev/null
+++ b/Documentation/i2c/DMA-considerations
@@ -0,0 +1,38 @@
+Linux I2C and DMA
+-----------------
+
+Given that I2C is a low-speed bus where largely small messages are transferred,
+it is not considered a prime user of DMA access. At this time of writing, only
+10% of I2C bus master drivers have DMA support implemented. And the vast
+majority of transactions are so small that setting up DMA for it will likely
+add more overhead than a plain PIO transfer.
+
+Therefore, it is *not* mandatory that the buffer of an I2C message is DMA safe.
+It does not seem reasonable to apply additional burdens when the feature is so
+rarely used. However, it is recommended to use a DMA-safe buffer if your
+message size is likely applicable for DMA. Most drivers have this threshold
+around 8 bytes. As of today, this is mostly an educated guess, however.
+
+To support this scenario, drivers wishing to implement DMA can use helper
+functions from the I2C core. One checks if a message is DMA capable in terms of
+size and memory type. It can optionally also create a bounce buffer:
+
+	i2c_check_msg_for_dma(msg, threshold, &bounce_buf);
+
+The bounce buffer handling from the core is generic and simple. It will always
+allocate a new bounce buffer. If you want a more sophisticated handling (e.g.
+reusing pre-allocated buffers), you can leave the pointer to the bounce buffer
+empty and implement your own handling based on the return value of the above
+function.
+
+The other helper function releases the bounce buffer. It ensures data is copied
+back to the message:
+
+	i2c_release_dma_bounce_buf(msg, bounce_buf);
+
+Please check the in-kernel documentation for details. The i2c-sh_mobile driver
+can be used as a reference example.
+
+If you plan to use DMA with I2C (or with any other bus, actually) make sure you
+have CONFIG_DMA_API_DEBUG enabled during development. It can help you find
+various issues which can be complex to debug otherwise.
-- 
2.11.0

[toc] | [next] | [standalone]

#1691173

Hi Wolfram,

On 2017-07-18 12:23:37 +0200, Wolfram Sang wrote:
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>

Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>

> ---
> Changes since v2:
> 
> * documentation updates. Hopefully better wording now
> 
>  Documentation/i2c/DMA-considerations | 38 ++++++++++++++++++++++++++++++++++++
>  1 file changed, 38 insertions(+)
>  create mode 100644 Documentation/i2c/DMA-considerations
> 
> diff --git a/Documentation/i2c/DMA-considerations b/Documentation/i2c/DMA-considerations
> new file mode 100644
> index 00000000000000..e46c24d65c8556
> --- /dev/null
> +++ b/Documentation/i2c/DMA-considerations
> @@ -0,0 +1,38 @@
> +Linux I2C and DMA
> +-----------------
> +
> +Given that I2C is a low-speed bus where largely small messages are transferred,
> +it is not considered a prime user of DMA access. At this time of writing, only
> +10% of I2C bus master drivers have DMA support implemented. And the vast
> +majority of transactions are so small that setting up DMA for it will likely
> +add more overhead than a plain PIO transfer.
> +
> +Therefore, it is *not* mandatory that the buffer of an I2C message is DMA safe.
> +It does not seem reasonable to apply additional burdens when the feature is so
> +rarely used. However, it is recommended to use a DMA-safe buffer if your
> +message size is likely applicable for DMA. Most drivers have this threshold
> +around 8 bytes. As of today, this is mostly an educated guess, however.
> +
> +To support this scenario, drivers wishing to implement DMA can use helper
> +functions from the I2C core. One checks if a message is DMA capable in terms of
> +size and memory type. It can optionally also create a bounce buffer:
> +
> +	i2c_check_msg_for_dma(msg, threshold, &bounce_buf);
> +
> +The bounce buffer handling from the core is generic and simple. It will always
> +allocate a new bounce buffer. If you want a more sophisticated handling (e.g.
> +reusing pre-allocated buffers), you can leave the pointer to the bounce buffer
> +empty and implement your own handling based on the return value of the above
> +function.
> +
> +The other helper function releases the bounce buffer. It ensures data is copied
> +back to the message:
> +
> +	i2c_release_dma_bounce_buf(msg, bounce_buf);
> +
> +Please check the in-kernel documentation for details. The i2c-sh_mobile driver
> +can be used as a reference example.
> +
> +If you plan to use DMA with I2C (or with any other bus, actually) make sure you
> +have CONFIG_DMA_API_DEBUG enabled during development. It can help you find
> +various issues which can be complex to debug otherwise.
> -- 
> 2.11.0
> 

-- 
Regards,
Niklas Söderlund

[toc] | [prev] | [next] | [standalone]

#1694297

On Tue, 18 Jul 2017 12:23:37 +0200
Wolfram Sang <wsa+renesas@sang-engineering.com> wrote:

> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Is this material not perhaps better placed in the sphinx docs?
Up to you of course as your subsystem ;)

Text is good though.

Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> ---
> Changes since v2:
> 
> * documentation updates. Hopefully better wording now
> 
>  Documentation/i2c/DMA-considerations | 38 ++++++++++++++++++++++++++++++++++++
>  1 file changed, 38 insertions(+)
>  create mode 100644 Documentation/i2c/DMA-considerations
> 
> diff --git a/Documentation/i2c/DMA-considerations b/Documentation/i2c/DMA-considerations
> new file mode 100644
> index 00000000000000..e46c24d65c8556
> --- /dev/null
> +++ b/Documentation/i2c/DMA-considerations
> @@ -0,0 +1,38 @@
> +Linux I2C and DMA
> +-----------------
> +
> +Given that I2C is a low-speed bus where largely small messages are transferred,
> +it is not considered a prime user of DMA access. At this time of writing, only
> +10% of I2C bus master drivers have DMA support implemented. And the vast
> +majority of transactions are so small that setting up DMA for it will likely
> +add more overhead than a plain PIO transfer.
> +
> +Therefore, it is *not* mandatory that the buffer of an I2C message is DMA safe.
> +It does not seem reasonable to apply additional burdens when the feature is so
> +rarely used. However, it is recommended to use a DMA-safe buffer if your
> +message size is likely applicable for DMA. Most drivers have this threshold
> +around 8 bytes. As of today, this is mostly an educated guess, however.
> +
> +To support this scenario, drivers wishing to implement DMA can use helper
> +functions from the I2C core. One checks if a message is DMA capable in terms of
> +size and memory type. It can optionally also create a bounce buffer:
> +
> +	i2c_check_msg_for_dma(msg, threshold, &bounce_buf);
> +
> +The bounce buffer handling from the core is generic and simple. It will always
> +allocate a new bounce buffer. If you want a more sophisticated handling (e.g.
> +reusing pre-allocated buffers), you can leave the pointer to the bounce buffer
> +empty and implement your own handling based on the return value of the above
> +function.
> +
> +The other helper function releases the bounce buffer. It ensures data is copied
> +back to the message:
> +
> +	i2c_release_dma_bounce_buf(msg, bounce_buf);
> +
> +Please check the in-kernel documentation for details. The i2c-sh_mobile driver
> +can be used as a reference example.
> +
> +If you plan to use DMA with I2C (or with any other bus, actually) make sure you
> +have CONFIG_DMA_API_DEBUG enabled during development. It can help you find
> +various issues which can be complex to debug otherwise.

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web