On 07/07/2017 08:21 PM, Ferruh Yigit wrote:
Document NIC features, add more information about them and add more
implementation related support.
Signed-off-by: Ferruh Yigit <ferruh.yi...@intel.com>
Signed-off-by: John McNamara <john.mcnam...@intel.com>
Reviewed-by: Andrew Rybchenko <arybche...@solarflare.com>
Few minor questions/notes below.
---
Cc: Olivier Matz <olivier.m...@6wind.com>
v4:
* Apply review comments from Andrew Rybchenko
* Add tags to the information, to clarify in, out, related data.
v3:
* received updates from John, Thanks!
v2:
* Add more details, mbuf and API fields
* Formatting added
TODO:
- Not all features all fully documented, need help from community
- Instead of having a new file, it would be nice to auto generate this
file from defaults.ini, and perhaps add extra information as comment to
that file.
- Some features are implementing eth_dev_ops and some are updates in
Rx/Tx path. These features can be grouped together.
- A link from overview file per feature would be nice.
---
doc/guides/nics/overview.rst | 4 +
doc/guides/nics/overview_features.rst | 858 ++++++++++++++++++++++++++++++++++
2 files changed, 862 insertions(+)
create mode 100644 doc/guides/nics/overview_features.rst
<...>
+.. _nic_features_rss_hash:
+
+RSS hash
+--------
+
+Supports RSS hashing on RX.
+
+* **[uses] user config**: ``dev_conf.rxmode.mq_mode`` =
``ETH_MQ_RX_RSS_FLAG``.
+* **[provides] mbuf**: ``mbuf.ol_flags:PKT_RX_RSS_H()ASH``, ``mbuf.rss``.
Is it indented that dev_info.flow_type_rss_offloads not mentioned here?
<...>
+
+.. _nic_features_inner_l3_checksum:
+
+Inner L3 checksum
+-----------------
+
+Supports inner packet L3 checksum.
+
+
+.. _nic_features_inner_l4_checksum:
+
+Inner L4 checksum
+-----------------
+
+Supports inner packet L4 checksum.
+
As I understand outer checksum (PKT_RX_EIP_CKSUM_BAD for Rx) and
IPv4/IPv6 flags and
offload capabilities should be mentioned here. If ourter flags are set,
it means that usual checksum
flags corresponds to inner packet.
<...>
+
+.. _nic_features_rx_descriptor_status:
+
+Rx descriptor status
+--------------------
+
+Supports check the status of a Rx descriptor. When ``rx_descriptor_status`` is
+used, status can be "Available", "Done" or "Unavailable". When
+``rx_descriptor_done`` is used, status can be "DD bit is set" or "DD bit is
+not set".
+
+* **[implements] eth_dev_ops**: ``rx_descriptor_status``.
+* **[related] API**: ``rte_eth_rx_descriptor_status()``.
+* **[implements] eth_dev_ops**: ``rx_descriptor_done``.
+* **[related] API**: ``rte_eth_rx_descriptor_done()``.
I'm not sure if _done API should be mentioned here. If yes, may be
status of the _done
API should be highlighted (as far as I remember it was ideas to remove it).
<...>
+
+.. _nic_features_bsd_nic_uio:
+
+BSD nic_uio
+-----------
+
+BSD ``nic_uio`` module supported.
+
+
+.. _nic_features_linux_uio:
+
+Linux UIO
+---------
+
+Works with ``igb_uio`` kernel module.
+
+
+.. _nic_features_linux_vfio:
+
+Linux VFIO
+----------
+
+Works with ``vfio-pci`` kernel module.
Should we mention that PMD provides RTE_PMD_REGISTER_KMOD_DEP with
corresponding kernel module listed.
<...>
