[v3] doc: update hairpin data buffer size config
Checks
Commit Message
This patch updates the MLX5 PMD and release notes documentations.
Adding the guideline for hairpin data buffer size configuration.
Signed-off-by: Bing Zhao <bingz@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
---
v2: update the range description of hairpin data buffer size
v3: correct the minimum value of the range
---
doc/guides/nics/mlx5.rst | 16 ++++++++++++++++
doc/guides/rel_notes/release_20_05.rst | 1 +
2 files changed, 17 insertions(+)
Comments
21/04/2020 09:29, Bing Zhao:
> This patch updates the MLX5 PMD and release notes documentations.
> Adding the guideline for hairpin data buffer size configuration.
>
> Signed-off-by: Bing Zhao <bingz@mellanox.com>
> Acked-by: Ori Kam <orika@mellanox.com>
> ---
> --- a/doc/guides/nics/mlx5.rst
> +++ b/doc/guides/nics/mlx5.rst
> +- ``hp_buf_log_sz`` parameter [int]
> +
> + The total data buffer size of a hairpin queue (logarithmic form), in bytes.
> + PMD will set the data buffer size to 2 ** ``hp_buf_log_sz``, both for RX & TX.
> + The capacity of the value is sepcified by the Firmware, and the initialization
sepcified -> specified
Firmware -> firmware
> + will get a failure if it is out of scope.
> + The range of the value is from 11 to 19 right now, and the supported frame
> + size of a single packet for hairpin is from 512B to 128KB. It might change if
> + different Firmware release is being used. By using a small value, it could
Firmware -> firmware
> + reduce the memory consumption but not work with large frame. If the value is
> + too large, the memory consumption will be high and some potential performance
> + degradation will be introduced.
> + By default PMD will set this value to 16, which mean that 9KB jumbo frames
By default, the PMD
> + will be supported.
> +
> --- a/doc/guides/rel_notes/release_20_05.rst
> +++ b/doc/guides/rel_notes/release_20_05.rst
> @@ -62,6 +62,7 @@ New Features
>
> * Added support for matching on IPv4 Time To Live and IPv6 Hop Limit.
> * Added support for creating Relaxed Ordering Memory Regions.
> + * Added support for configuring Haripin queue data buffer size.
Haripin -> Hairpin
> -----Original Message-----
> From: Thomas Monjalon <thomas@monjalon.net>
> Sent: Wednesday, April 22, 2020 10:13 AM
> To: Bing Zhao <bingz@mellanox.com>
> Cc: john.mcnamara@intel.com; marko.kovacevic@intel.com; Slava
> Ovsiienko <viacheslavo@mellanox.com>; Ori Kam
> <orika@mellanox.com>; Matan Azrad <matan@mellanox.com>;
> Shahaf Shuler <shahafs@mellanox.com>; Raslan Darawsheh
> <rasland@mellanox.com>; dev@dpdk.org
> Subject: Re: [PATCH v3] doc: update hairpin data buffer size config
>
> 21/04/2020 09:29, Bing Zhao:
> > This patch updates the MLX5 PMD and release notes
> documentations.
> > Adding the guideline for hairpin data buffer size configuration.
> >
> > Signed-off-by: Bing Zhao <bingz@mellanox.com>
> > Acked-by: Ori Kam <orika@mellanox.com>
> > ---
> > --- a/doc/guides/nics/mlx5.rst
> > +++ b/doc/guides/nics/mlx5.rst
> > +- ``hp_buf_log_sz`` parameter [int]
> > +
> > + The total data buffer size of a hairpin queue (logarithmic form), in
> bytes.
> > + PMD will set the data buffer size to 2 ** ``hp_buf_log_sz``, both
> for RX & TX.
> > + The capacity of the value is sepcified by the Firmware, and the
> > + initialization
>
> sepcified -> specified
> Firmware -> firmware
>
> > + will get a failure if it is out of scope.
> > + The range of the value is from 11 to 19 right now, and the
> > + supported frame size of a single packet for hairpin is from 512B to
> > + 128KB. It might change if different Firmware release is being used.
> > + By using a small value, it could
>
> Firmware -> firmware
>
> > + reduce the memory consumption but not work with large frame. If
> the
> > + value is too large, the memory consumption will be high and some
> > + potential performance degradation will be introduced.
> > + By default PMD will set this value to 16, which mean that 9KB
> jumbo
> > + frames
>
> By default, the PMD
>
> > + will be supported.
> > +
>
> > --- a/doc/guides/rel_notes/release_20_05.rst
> > +++ b/doc/guides/rel_notes/release_20_05.rst
> > @@ -62,6 +62,7 @@ New Features
> >
> > * Added support for matching on IPv4 Time To Live and IPv6 Hop
> Limit.
> > * Added support for creating Relaxed Ordering Memory Regions.
> > + * Added support for configuring Haripin queue data buffer size.
>
> Haripin -> Hairpin
>
>
Many thanks, Thomas. I corrected them and sent the v4.
@@ -99,6 +99,7 @@ Features
- Support for multiple rte_flow groups.
- Per packet no-inline hint flag to disable packet data copying into Tx descriptors.
- Hardware LRO.
+- Hairpin.
Limitations
-----------
@@ -786,6 +787,21 @@ Run-time configuration
If this parameter is not specified, by default PMD will set
the smallest value supported by HW.
+- ``hp_buf_log_sz`` parameter [int]
+
+ The total data buffer size of a hairpin queue (logarithmic form), in bytes.
+ PMD will set the data buffer size to 2 ** ``hp_buf_log_sz``, both for RX & TX.
+ The capacity of the value is sepcified by the Firmware, and the initialization
+ will get a failure if it is out of scope.
+ The range of the value is from 11 to 19 right now, and the supported frame
+ size of a single packet for hairpin is from 512B to 128KB. It might change if
+ different Firmware release is being used. By using a small value, it could
+ reduce the memory consumption but not work with large frame. If the value is
+ too large, the memory consumption will be high and some potential performance
+ degradation will be introduced.
+ By default PMD will set this value to 16, which mean that 9KB jumbo frames
+ will be supported.
+
.. _mlx5_firmware_config:
Firmware configuration
@@ -62,6 +62,7 @@ New Features
* Added support for matching on IPv4 Time To Live and IPv6 Hop Limit.
* Added support for creating Relaxed Ordering Memory Regions.
+ * Added support for configuring Haripin queue data buffer size.
* Added support for jumbo frame size (9K MTU) in Multi-Packet RQ mode.
* Optimized the memory consumption of flow.