dpdk.git
3 years agonet/iavf: support CRC strip disabling
Haiyue Wang [Tue, 12 Jan 2021 08:13:00 +0000 (16:13 +0800)]
net/iavf: support CRC strip disabling

The VF will check the PF's CRC strip capability firstly, then set the
'CRC strip disable' value in the queue configuration according to the
RX CRC offload setting.

Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agocommon/iavf: update copyright date
Qi Zhang [Thu, 7 Jan 2021 04:53:33 +0000 (12:53 +0800)]
common/iavf: update copyright date

Updated the Copyright for 2021.
Updated FreeBSD IAVF driver of version.

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agocommon/iavf: support VLAN offload by DCF
Qi Zhang [Thu, 7 Jan 2021 05:15:04 +0000 (13:15 +0800)]
common/iavf: support VLAN offload by DCF

Add new opcode VIRTCHNL_OP_DCF_VLAN_OFFLOAD to set VLAN offload
by DCF, the virtchnl message includes:
1. A valid target VF
2. Type of VLAN to be supported: outer or inner
3. Ethertype of the VLAN (either 0x8100 or 0x88A8 or 0x9100)
4. VLAN insert settings
   a). No insert offload, VLAN ID in the packet (default)
   b). Offload via transmit descriptor
   c). Insert as a port VLAN (via VSI)
5. VLAN strip settings
   a). Strip (and discard)
   b). Strip and place in descriptor
   c). No Strip
6. VLAN ID for the target VF

Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agocommon/iavf: support new VLAN capabilities
Qi Zhang [Thu, 7 Jan 2021 05:07:08 +0000 (13:07 +0800)]
common/iavf: support new VLAN capabilities

Currently VIRTCHNL only allows for VLAN filtering and offloads to happen
on a single 802.1Q VLAN. Add support to filter and offload on inner,
outer, and/or inner + outer VLANs.

This is done by introducing the new capability
VIRTCHNL_VF_OFFLOAD_VLAN_V2. The flow to negotiate this new capability
is shown below.

1. VF - sets the VIRTCHNL_VF_OFFLOAD_VLAN_V2 bit in the
   virtchnl_vf_resource.vf_caps_flags during the
   VIRTCHNL_OP_GET_VF_RESOURCES request message. The VF should also set
   the VIRTCHNL_VF_OFFLOAD_VLAN bit in case the PF driver doesn't
   support the new capability.

2. PF - sets the VLAN capability bit it supports in the
   VIRTCHNL_OP_GET_VF_RESOURCES response message. This will either be
   VIRTCHNL_VF_OFFLOAD_VLAN_V2, VIRTCHNL_VF_OFFLOAD_VLAN, or none.

3. VF - If the VIRTCHNL_VF_OFFLOAD_VLAN_V2 capability was ACK'd by the
   PF, then the VF needs to request the VLAN capabilities of the
   PF/Device by issuing a VIRTCHNL_OP_GET_OFFLOAD_VLAN_V2_CAPS request.
   If the VIRTCHNL_VF_OFFLOAD_VLAN capability was ACK'd then the VF
   knows only single 802.1Q VLAN filtering/offloads are supported. If no
   VLAN capability is ACK'd then the PF/Device doesn't support hardware
   VLAN filtering/offloads for this VF.

4. PF - Populates the virtchnl_vlan_caps structure based on what it
   allows/supports for that VF and sends that response via
   VIRTCHNL_OP_GET_OFFLOAD_VLAN_V2_CAPS.

After VIRTCHNL_OP_GET_OFFLOAD_VLAN_V2_CAPS is successfully negotiated
the VF driver needs to interpret the capabilities supported by the
underlying PF/Device. The VF will be allowed to filter/offload the
inner 802.1Q, outer (various ethertype), inner 802.1Q + outer
(various ethertypes), or none based on which fields are set.

The VF will also need to interpret where the VLAN tag should be inserted
and/or stripped based on the negotiated capabilities.

Also, update the virtchnl_op_str() function to support the added opcodes.

Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice: enlarge Rx queue rearm threshold to 64
Leyi Rong [Wed, 6 Jan 2021 05:35:48 +0000 (13:35 +0800)]
net/ice: enlarge Rx queue rearm threshold to 64

We observe performance drop on ice AVX512 data path after stop and
start by using testpmd.

As CPU polling is faster in AVX512 path, L3 contested accesses is
intensified when rxrearm_start is a random value after testpmd
stop/start.

Enlarge ICE_RXQ_REARM_THRESH to 64 to ease the contested accesses and
fix the performance drop issue.

Cc: stable@dpdk.org
Signed-off-by: Leyi Rong <leyi.rong@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice: disable IPv4 checksum offload in vector Tx
Murphy Yang [Fri, 8 Jan 2021 07:17:52 +0000 (07:17 +0000)]
net/ice: disable IPv4 checksum offload in vector Tx

ICE choices vector TX path or basic TX path by macro
'ICE_NO_VECTOR_FLAGS'.

This patch adds 'DEV_TX_OFFLOAD_IPV4_CKSUM' in 'ICE_NO_VECTOR_FLAGS'
to make IPv4 checksum offload processed by basic TX path.

Fixes: a22483208800 ("net/ice: disable TSO offload in vector path")
Cc: stable@dpdk.org
Signed-off-by: Murphy Yang <murphyx.yang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice: fix RSS lookup table initialization
Alvin Zhang [Fri, 8 Jan 2021 07:29:10 +0000 (15:29 +0800)]
net/ice: fix RSS lookup table initialization

RSS look-up table initialization is done incorrectly due to
divide-by-zero error.
Add a check to rx-queue count.

Fixes: 50370662b727 ("net/ice: support device and queue ops")
Cc: stable@dpdk.org
Signed-off-by: Alvin Zhang <alvinx.zhang@intel.com>
Tested-by: Wei Xie <weix.xie@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/iavf: fix conflicting RSS combination rules
Murphy Yang [Thu, 7 Jan 2021 09:17:10 +0000 (09:17 +0000)]
net/iavf: fix conflicting RSS combination rules

Currently, when use 'flow' command to create a rule with following
invalid RSS type combination, it can be created successfully.

Invalid RSS combinations list:
 - ETH_RSS_IPV4 | ETH_RSS_NONFRAG_IPV4_TCP
 - ETH_RSS_IPV6 | ETH_RSS_NONFRAG_IPV6_TCP

This patch adds these combinations in 'invalid_rss_comb' array to do
valid check, if the combination check failed, the rule will be created
unsuccessful.

Fixes: 91f27b2e39ab ("net/iavf: refactor RSS")
Cc: stable@dpdk.org
Signed-off-by: Murphy Yang <murphyx.yang@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
3 years agonet/txgbe: add security type in flow action
Jiawen Wu [Fri, 18 Dec 2020 09:37:02 +0000 (17:37 +0800)]
net/txgbe: add security type in flow action

Add security type in flow action.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: add security offload in Rx and Tx
Jiawen Wu [Fri, 18 Dec 2020 09:37:01 +0000 (17:37 +0800)]
net/txgbe: add security offload in Rx and Tx

Add security offload in Rx and Tx process.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: destroy security session
Jiawen Wu [Fri, 18 Dec 2020 09:37:00 +0000 (17:37 +0800)]
net/txgbe: destroy security session

Add support to clear a security session's private data,
get the size of a security session,
add update the mbuf with provided metadata.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: add security session create operation
Jiawen Wu [Fri, 18 Dec 2020 09:36:59 +0000 (17:36 +0800)]
net/txgbe: add security session create operation

Add support to configure a security session.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: add IPsec context creation
Jiawen Wu [Fri, 18 Dec 2020 09:36:58 +0000 (17:36 +0800)]
net/txgbe: add IPsec context creation

Initialize securiry context, and add support to get
security capabilities.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: add TM hierarchy commit
Jiawen Wu [Fri, 18 Dec 2020 09:36:56 +0000 (17:36 +0800)]
net/txgbe: add TM hierarchy commit

Add traffic manager hierarchy commit.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: support TM node add and delete
Jiawen Wu [Fri, 18 Dec 2020 09:36:55 +0000 (17:36 +0800)]
net/txgbe: support TM node add and delete

Support traffic manager node add and delete operations.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: support TM shaper profile add and delete
Jiawen Wu [Fri, 18 Dec 2020 09:36:54 +0000 (17:36 +0800)]
net/txgbe: support TM shaper profile add and delete

Support traffic manager profile add and delete operations.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: add TM capabilities get operation
Jiawen Wu [Fri, 18 Dec 2020 09:36:53 +0000 (17:36 +0800)]
net/txgbe: add TM capabilities get operation

Add support to get traffic manager capabilities.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: add TM configuration init and uninit
Jiawen Wu [Fri, 18 Dec 2020 09:36:52 +0000 (17:36 +0800)]
net/txgbe: add TM configuration init and uninit

Add traffic manager configuration init and uninit operations.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: support UDP tunnel port add and delete
Jiawen Wu [Fri, 18 Dec 2020 09:36:51 +0000 (17:36 +0800)]
net/txgbe: support UDP tunnel port add and delete

Support UDP tunnel port add and delete operations.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: flush all filters
Jiawen Wu [Fri, 18 Dec 2020 09:36:50 +0000 (17:36 +0800)]
net/txgbe: flush all filters

Add support to flush all the filters.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: support destroying consistent filter
Jiawen Wu [Fri, 18 Dec 2020 09:36:49 +0000 (17:36 +0800)]
net/txgbe: support destroying consistent filter

Add a function to destroy the flow filter.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: support creating consistent filter
Jiawen Wu [Fri, 18 Dec 2020 09:36:48 +0000 (17:36 +0800)]
net/txgbe: support creating consistent filter

Create a flow rule, to use the matched filter which the rule hit first.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: parse RSS filter
Jiawen Wu [Fri, 18 Dec 2020 09:36:47 +0000 (17:36 +0800)]
net/txgbe: parse RSS filter

Check if the rule is a RSS filter rule, and get the RSS info.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: restore RSS filter
Jiawen Wu [Fri, 18 Dec 2020 09:36:46 +0000 (17:36 +0800)]
net/txgbe: restore RSS filter

Add support to restore RSS filter.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: parse flow director filter
Jiawen Wu [Fri, 18 Dec 2020 09:36:45 +0000 (17:36 +0800)]
net/txgbe: parse flow director filter

Check if the rule is a flow director rule, and get the flow director info.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: support flow director filter add and delete
Jiawen Wu [Fri, 18 Dec 2020 09:36:44 +0000 (17:36 +0800)]
net/txgbe: support flow director filter add and delete

Support add and delete operations on flow director filter.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: configure flow director filter
Jiawen Wu [Fri, 18 Dec 2020 09:36:43 +0000 (17:36 +0800)]
net/txgbe: configure flow director filter

Configure flow director filter with it enabled.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: add flow director filter init and uninit
Jiawen Wu [Fri, 18 Dec 2020 09:36:42 +0000 (17:36 +0800)]
net/txgbe: add flow director filter init and uninit

Add flow director filter init and uninit operations.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: parse L2 tunnel filter
Jiawen Wu [Fri, 18 Dec 2020 09:36:41 +0000 (17:36 +0800)]
net/txgbe: parse L2 tunnel filter

Check if the rule is a L2 tunnel rule, and get the L2 tunnel info.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: support L2 tunnel filter add and delete
Jiawen Wu [Fri, 18 Dec 2020 09:36:40 +0000 (17:36 +0800)]
net/txgbe: support L2 tunnel filter add and delete

Support L2 tunnel filter add and delete operations.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: config L2 tunnel filter with e-tag
Jiawen Wu [Fri, 18 Dec 2020 09:36:39 +0000 (17:36 +0800)]
net/txgbe: config L2 tunnel filter with e-tag

Config L2 tunnel filter with e-tag.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: add L2 tunnel filter init and uninit
Jiawen Wu [Fri, 18 Dec 2020 09:36:38 +0000 (17:36 +0800)]
net/txgbe: add L2 tunnel filter init and uninit

Add L2 tunnel filter init and uninit.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: parse syn filter
Jiawen Wu [Fri, 18 Dec 2020 09:36:37 +0000 (17:36 +0800)]
net/txgbe: parse syn filter

Check if the rule is a TCP SYN rule, and get the SYN info.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: support syn filter add and delete
Jiawen Wu [Fri, 18 Dec 2020 09:36:36 +0000 (17:36 +0800)]
net/txgbe: support syn filter add and delete

Support add and delete operations on syn filter.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: parse ethertype filter
Jiawen Wu [Fri, 18 Dec 2020 09:36:35 +0000 (17:36 +0800)]
net/txgbe: parse ethertype filter

Check if the rule is a ethertype rule, and get the ethertype info.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: support ethertype filter add and delete
Jiawen Wu [Fri, 18 Dec 2020 09:36:34 +0000 (17:36 +0800)]
net/txgbe: support ethertype filter add and delete

Support add and delete operations on ethertype filter.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: parse n-tuple filter
Jiawen Wu [Fri, 18 Dec 2020 09:36:33 +0000 (17:36 +0800)]
net/txgbe: parse n-tuple filter

Check if the rule is a n-tuple rule, and get the n-tuple info.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: support ntuple filter add and delete
Jiawen Wu [Fri, 18 Dec 2020 09:36:32 +0000 (17:36 +0800)]
net/txgbe: support ntuple filter add and delete

Support add and delete operations on ntuple filter.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: add ntuple filter init and uninit
Jiawen Wu [Fri, 18 Dec 2020 09:36:31 +0000 (17:36 +0800)]
net/txgbe: add ntuple filter init and uninit

Add ntuple filter init and uninit.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/txgbe: add generic flow API
Jiawen Wu [Fri, 18 Dec 2020 09:36:30 +0000 (17:36 +0800)]
net/txgbe: add generic flow API

Introduce rte_flow with its validate, create, destroy and flush
operations into txgbe PMD.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
3 years agonet/ionic: stop queues when LIF is stopped
Andrew Boyer [Wed, 16 Dec 2020 21:12:57 +0000 (13:12 -0800)]
net/ionic: stop queues when LIF is stopped

Otherwise they cannot be restarted, because the FW will reject INIT
or ENA commands on queues which are already running.

Signed-off-by: Andrew Boyer <aboyer@pensando.io>
3 years agonet/ionic: improve queue state handling
Andrew Boyer [Wed, 16 Dec 2020 21:12:56 +0000 (13:12 -0800)]
net/ionic: improve queue state handling

Skip ionic_lif_[rxq|txq]_init() in queue start if it's already done.
Move ionic_lif_[rxq|txq]_deinit() from queue stop to queue release.

This allows the queues to be restarted.

Signed-off-by: Andrew Boyer <aboyer@pensando.io>
3 years agonet/ionic: improve link state handling
Andrew Boyer [Wed, 16 Dec 2020 21:12:55 +0000 (13:12 -0800)]
net/ionic: improve link state handling

Add UP and FW_RESET state flags.
Update the stack info when the link state changes.
Convert set_link_up/set_link_down to lif_start/lif_stop.
Condition reported link state on UP flag.

Signed-off-by: Andrew Boyer <aboyer@pensando.io>
3 years agonet/ionic: complete release on close
Andrew Boyer [Wed, 16 Dec 2020 21:12:54 +0000 (13:12 -0800)]
net/ionic: complete release on close

ionic_dev_close() is responsible for destroying the ethdev, lif, and
adapter. eth_ionic_dev_remove() calls ionic_dev_close().

Remove-on-close is now required behavior for a PMD.
Remove the UNMAINTAINED flag.

Signed-off-by: Andrew Boyer <aboyer@pensando.io>
3 years agonet/ionic: remove multi-LIF support
Andrew Boyer [Wed, 16 Dec 2020 21:12:53 +0000 (13:12 -0800)]
net/ionic: remove multi-LIF support

This feature is unused, so remove it.

There is exactly one adapter / lif / ethdev per port.

Signed-off-by: Andrew Boyer <aboyer@pensando.io>
3 years agonet/ionic: preserve Rx mode across LIF stop/start
Andrew Boyer [Wed, 16 Dec 2020 21:12:52 +0000 (13:12 -0800)]
net/ionic: preserve Rx mode across LIF stop/start

Otherwise, non-default settings (like PROMISC) get reset.
This will become important when link toggling is tied to LIF stop/start.

Signed-off-by: Andrew Boyer <aboyer@pensando.io>
3 years agonet/ionic: preserve RSS state unless RETA size changes
Andrew Boyer [Wed, 16 Dec 2020 21:12:51 +0000 (13:12 -0800)]
net/ionic: preserve RSS state unless RETA size changes

This preserves settings across a LIF stop/start.
This will become important when link toggling is tied to LIF stop/start.

Signed-off-by: Andrew Boyer <aboyer@pensando.io>
3 years agovhost: enhance async enqueue for small packets
Jiayu Hu [Mon, 11 Jan 2021 12:16:27 +0000 (07:16 -0500)]
vhost: enhance async enqueue for small packets

Async enqueue offloads large copies to DMA devices, and small copies
are still performed by the CPU. However, it requires users to get
enqueue completed packets by rte_vhost_poll_enqueue_completed(), even
if they are completed by the CPU when rte_vhost_submit_enqueue_burst()
returns. This design incurs extra overheads of tracking completed
pktmbufs and function calls, thus degrading performance on small packets.

This patch enhances async enqueue for small packets by enabling
rte_vhost_submit_enqueue_burst() to return completed packets.

Signed-off-by: Jiayu Hu <jiayu.hu@intel.com>
Tested-by: Yinan Wang <yinan.wang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
3 years agovhost: cleanup async enqueue
Jiayu Hu [Mon, 11 Jan 2021 12:16:26 +0000 (07:16 -0500)]
vhost: cleanup async enqueue

This patch removes unnecessary check and function calls, and it changes
appropriate types for internal variables and fixes typos.

Signed-off-by: Jiayu Hu <jiayu.hu@intel.com>
Tested-by: Yinan Wang <yinan.wang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
3 years agonet/virtio: fix memory init with vDPA backend
Maxime Coquelin [Fri, 8 Jan 2021 09:41:49 +0000 (10:41 +0100)]
net/virtio: fix memory init with vDPA backend

This patch fixes an overhead met with mlx5-vdpa Kernel
driver, where for every page in the mapped area, all the
memory tables gets updated. For example, with 2MB hugepages,
a single IOTLB_UPDATE for a 1GB region causes 512 memory
updates on mlx5-vdpa side.

Using batching mode, the mlx5 driver will only trigger a
single memory update for all the IOTLB updates that happen
between the batch begin and batch end commands.

Fixes: 6b901437056e ("net/virtio: introduce vhost-vDPA backend")
Cc: stable@dpdk.org
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
3 years agonet/virtio: add missing backend features negotiation
Maxime Coquelin [Fri, 8 Jan 2021 09:41:48 +0000 (10:41 +0100)]
net/virtio: add missing backend features negotiation

This patch adds missing backend features negotiation for
in Vhost-vDPA. Without it, IOTLB messages v2 could be sent
by Virtio-user PMD while not supported by the backend.

Fixes: 6b901437056e ("net/virtio: introduce vhost-vDPA backend")
Cc: stable@dpdk.org
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
3 years agonet/cxgbe: accept VLAN flow items without ethertype
Karra Satwik [Sun, 20 Dec 2020 22:44:43 +0000 (04:14 +0530)]
net/cxgbe: accept VLAN flow items without ethertype

When apps pass the RTE_FLOW_ITEM_TYPE_VLAN without setting the
ethertype field in RTE_FLOW_ITEM_TYPE_ETH, then assume 0x8100
VLAN by default and don't reject the rule.

Fixes: 55f003d8884c ("net/cxgbe: support flow API for matching QinQ VLAN")
Cc: stable@dpdk.org
Signed-off-by: Karra Satwik <kaara.satwik@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
3 years agonet/enic: remove deprecated flow director code
John Daley [Thu, 17 Dec 2020 01:37:15 +0000 (17:37 -0800)]
net/enic: remove deprecated flow director code

The Flow Director (FDIR) API was removed in release 20.11.
This patch removes the remainder of the FDIR code in the
PMD.

Signed-off-by: John Daley <johndale@cisco.com>
Reviewed-by: Hyong Youb Kim <hyonkim@cisco.com>
3 years agonet/axgbe: support reading FW version
Selwin Sebastian [Wed, 6 Jan 2021 08:00:58 +0000 (13:30 +0530)]
net/axgbe: support reading FW version

Added support for fw_version_get API

Signed-off-by: Selwin Sebastian <selwin.sebastian@amd.com>
Acked-by: Somalapuram Amaranath <asomalap@amd.com>
3 years agonet/ice: refactor PF RSS
Xuan Ding [Fri, 8 Jan 2021 08:38:44 +0000 (08:38 +0000)]
net/ice: refactor PF RSS

This patch refactors the PF RSS code based on the below design:
1. ice_pattern_match_item->input_set_mask is the superset of
   ETH_RSS_xxx.
2. ice_pattern_match_item->meta is the ice_rss_hash_cfg template.
3. ice_hash_parse_pattern will generate pattern hint.
4. ice_hash_parse_action will refine the ice_rss_hash_cfg based on
   the pattern hint and rss_type.
5. The refine process includes:
   1)  refine protocol headers(VLAN/PPPOE/GTPU).
   2)  refine hash bit fields of l2, l3, l4.
   3)  refine hash bit fields for gtpu header.

Signed-off-by: Xuan Ding <xuan.ding@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agoconfig/arm: add Neoverse N2
Ruifeng Wang [Tue, 12 Jan 2021 02:57:08 +0000 (02:57 +0000)]
config/arm: add Neoverse N2

Add Arm Neoverse N2 cpu support.

Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
3 years agocommon/octeontx2: fix build with SVE
Ruifeng Wang [Tue, 12 Jan 2021 02:57:07 +0000 (02:57 +0000)]
common/octeontx2: fix build with SVE

Building with gcc 10.2 with SVE extension enabled got error:

{standard input}: Assembler messages:
{standard input}:4002: Error: selected processor does not support `mov z3.b,#0'
{standard input}:4003: Error: selected processor does not support `whilelo p1.b,xzr,x7'
{standard input}:4005: Error: selected processor does not support `ld1b z0.b,p1/z,[x8]'
{standard input}:4006: Error: selected processor does not support `whilelo p4.s,wzr,w7'

This is because inline assembly code explicitly resets cpu model to
not have SVE support. Thus SVE instructions generated by compiler
auto vectorization got rejected by assembler.

Added SVE to the cpu model specified by inline assembly for SVE support.
Not replacing the inline assembly with C atomics because the driver relies
on specific LSE instruction to interface to co-processor [1].

Fixes: 8a4f835971f5 ("common/octeontx2: add IO handling APIs")
Cc: stable@dpdk.org
[1] https://mails.dpdk.org/archives/dev/2021-January/196092.html

Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Jerin Jacob <jerinj@marvell.com>
3 years agonet/octeontx: fix build with SVE
Ruifeng Wang [Tue, 12 Jan 2021 02:57:06 +0000 (02:57 +0000)]
net/octeontx: fix build with SVE

Building with gcc 10.2 with SVE extension enabled got error:

{standard input}: Assembler messages:
{standard input}:91: Error: selected processor does not support `addvl x4,x8,#-1'
{standard input}:95: Error: selected processor does not support `ptrue p1.d,all'
{standard input}:135: Error: selected processor does not support `whilelo p2.d,xzr,x5'
{standard input}:137: Error: selected processor does not support `decb x1'

This is because inline assembly code explicitly resets cpu model to
not have SVE support. Thus SVE instructions generated by compiler
auto vectorization got rejected by assembler.

Added SVE to the cpu model specified by inline assembly for SVE support.
Not replacing the inline assembly with C atomics because the driver relies
on specific LSE instruction to interface to co-processor [1].

Fixes: f0c7bb1bf778 ("net/octeontx/base: add octeontx IO operations")
Cc: stable@dpdk.org
[1] https://mails.dpdk.org/archives/dev/2021-January/196092.html

Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Jerin Jacob <jerinj@marvell.com>
3 years agonet/hns3: fix build with SVE
Ruifeng Wang [Tue, 12 Jan 2021 02:57:05 +0000 (02:57 +0000)]
net/hns3: fix build with SVE

Building with SVE extension enabled stopped with error:

 error: ACLE function â€˜svwhilelt_b64_s32’ requires ISA extension â€˜sve’
   18 | #define PG64_256BIT  svwhilelt_b64(0, 4)

This is caused by unintentional cflags reset.
Fixed the issue by not touching cflags, and using flags defined by
compiler.

Fixes: 952ebacce4f2 ("net/hns3: support SVE Rx")
Cc: stable@dpdk.org
Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
3 years agolpm/arm: support SVE
Ruifeng Wang [Tue, 12 Jan 2021 02:57:04 +0000 (02:57 +0000)]
lpm/arm: support SVE

Added new path to do lpm4 lookup by using scalable vector extension.
The SVE path will be selected if compiler has flag SVE set.

Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Acked-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
3 years agotest: improve coverage on LPM tbl8
Ruifeng Wang [Thu, 14 Jan 2021 06:59:25 +0000 (06:59 +0000)]
test: improve coverage on LPM tbl8

Existing test cases create 256 tbl8 groups for testing. The number covers
only 8 bit next_hop/group field. Since the next_hop/group field had been
extended to 24-bits, creating more than 256 groups in tests can improve
the coverage.

Coverage was not expanded to reach the max supported group number, because
it would take too much time to run for this fast-test.

Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Tested-by: David Christensen <drc@linux.vnet.ibm.com>
Acked-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
3 years agolpm: fix vector IPv4 lookup
Ruifeng Wang [Thu, 14 Jan 2021 06:59:22 +0000 (06:59 +0000)]
lpm: fix vector IPv4 lookup

rte_lpm_lookupx4 could return wrong next hop when more than 256 tbl8
groups are created. This is caused by incorrect type casting of tbl8
group index that been stored in tbl24 entry. The casting caused group
index truncation and hence wrong tbl8 group been searched.

Issue fixed by applying proper mask to tbl24 entry to get tbl8 group index.

Fixes: dc81ebbacaeb ("lpm: extend IPv4 next hop field")
Fixes: cbc2f1dccfba ("lpm/arm: support NEON")
Fixes: d2cc7959342b ("lpm: add AltiVec for ppc64")
Cc: stable@dpdk.org
Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Tested-by: David Christensen <drc@linux.vnet.ibm.com>
Acked-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
3 years agofib6: improve AVX512 lookup performance
Vladimir Medvedkin [Tue, 15 Dec 2020 18:25:19 +0000 (18:25 +0000)]
fib6: improve AVX512 lookup performance

Improved performance for AVX512 FIB6 lookup by doubling the number
of flows being processed

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Ray Kinsella <mdr@ashroe.eu>
3 years agobuild: fix linker flags on Windows
Dmitry Kozlyuk [Tue, 12 Jan 2021 00:36:02 +0000 (03:36 +0300)]
build: fix linker flags on Windows

The --export-dynamic linker option is only applicable to ELF.
On Windows, where COFF is used, it causes warnings:

    x86_64-w64-mingw32-ld: warning: --export-dynamic is not supported
    for PE+ targets, did you mean --export-all-symbols? (MinGW)

    LINK : warning LNK4044: unrecognized option '/-export-dynamic';
    ignored (clang)

Don't add --export-dynamic on Windows anywhere.

Fixes: b031e13d7f0d ("build: fix plugin load on static build")
Cc: stable@dpdk.org
Signed-off-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
Acked-by: Ranjit Menon <ranjit.menon@intel.com>
3 years agodoc: add vtune profiling config to prog guide
Eugeny Parshutin [Wed, 2 Dec 2020 17:48:06 +0000 (20:48 +0300)]
doc: add vtune profiling config to prog guide

Return back 'profiling with vtune' section to profiling programmers
guide with updated instruction on how to enable vtune profiling
with meson configuration option.

Fixes: 89c67ae2cba7 ("doc: remove references to make from prog guide")
Cc: stable@dpdk.org
Signed-off-by: Eugeny Parshutin <eugeny.parshutin@linux.intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
3 years agodevtools: adjust verbosity of ABI check
Thomas Monjalon [Wed, 2 Dec 2020 17:15:21 +0000 (18:15 +0100)]
devtools: adjust verbosity of ABI check

The scripts gen-abi.sh and check-abi.sh are updated
to print error messages to stderr so they are likely never ignored.

When called from test-meson-builds.sh, the standard messages on stdout
can be more quiet depending on the verbosity settings.
The beginning of the ABI check is announced in verbose mode.
The commands are printed in very verbose mode.
The check result details are available in verbose mode.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
3 years agoapp/regex: measure performance with precise clock
Ophir Munk [Sun, 10 Jan 2021 11:10:23 +0000 (11:10 +0000)]
app/regex: measure performance with precise clock

Performance measurement (elapsed time and Gbps) are based on Linux
clock() API. The resolution is improved by replacing the clock() API
with rte_rdtsc_precise() API.

Signed-off-by: Ophir Munk <ophirmu@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
3 years agoapp/regex: measure performance per queue pair
Ophir Munk [Sun, 10 Jan 2021 11:10:22 +0000 (11:10 +0000)]
app/regex: measure performance per queue pair

Up to this commit measuring the parsing elapsed time and Giga bits per
second performance was done on the aggregation of all QPs (per core).
This commit separates the time measurements per individual QP.

Signed-off-by: Ophir Munk <ophirmu@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
3 years agoapp/regex: support multiple cores
Ophir Munk [Sun, 10 Jan 2021 11:10:21 +0000 (11:10 +0000)]
app/regex: support multiple cores

Up to this commit the regex application was running with multiple QPs on
a single core.  This commit adds the option to specify a number of cores
on which multiple QPs will run.
A new parameter 'nb_lcores' was added to configure the number of cores:
--nb_lcores <num of cores>.
If not configured the number of cores is set to 1 by default.  On
application startup a few initial steps occur by the main core: the
number of QPs and cores are parsed.  The QPs are distributed as evenly
as possible on the cores.  The regex device and all QPs are initialized.
The data file is read and saved in a buffer. Then for each core the
application calls rte_eal_remote_launch() with the worker routine
(run_regex) as its parameter.

Signed-off-by: Ophir Munk <ophirmu@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
3 years agoapp/regex: read data file once at startup
Ophir Munk [Sun, 10 Jan 2021 11:10:20 +0000 (11:10 +0000)]
app/regex: read data file once at startup

Up to this commit the input data file was read from scratch for each QP,
which is redundant. Starting from this commit the data file is read only
once at startup. Each QP will clone the data.

Signed-off-by: Ophir Munk <ophirmu@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
3 years agoapp/regex: support multiple queue pairs
Ophir Munk [Sun, 10 Jan 2021 11:10:19 +0000 (11:10 +0000)]
app/regex: support multiple queue pairs

Up to this commit the regex application used one QP which was assigned a
number of jobs, each with a different segment of a file to parse.  This
commit adds support for multiple QPs assignments. All QPs will be
assigned the same number of jobs, with the same segments of file to
parse. It will enable comparing functionality with different numbers of
QPs. All queues are managed on one core with one thread. This commit
focuses on changing routines API to support multi QPs, mainly, QP scalar
variables are replaced by per-QP struct instance.  The enqueue/dequeue
operations are interleaved as follows:
 enqueue(QP #1)
 enqueue(QP #2)
 ...
 enqueue(QP #n)
 dequeue(QP #1)
 dequeue(QP #2)
 ...
 dequeue(QP #n)

A new parameter 'nb_qps' was added to configure the number of QPs:
 --nb_qps <num of qps>.
If not configured, nb_qps is set to 1 by default.

Signed-off-by: Ophir Munk <ophirmu@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
3 years agoapp/regex: move mempool creation to worker routine
Ophir Munk [Sun, 10 Jan 2021 11:10:18 +0000 (11:10 +0000)]
app/regex: move mempool creation to worker routine

Function rte_pktmbuf_pool_create() is moved from init_port() routine to
run_regex() routine. Looking forward on multi core support - init_port()
will be called only once as part of application startup while mem pool
creation should be called multiple times (per core).

Signed-off-by: Ophir Munk <ophirmu@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
3 years agoregex/mlx5: add response flags
Ori Kam [Thu, 17 Dec 2020 10:37:31 +0000 (12:37 +0200)]
regex/mlx5: add response flags

This commit propagate the response flags from the regex engine.

Signed-off-by: Francis Kelly <fkelly@nvidia.com>
Signed-off-by: Ori Kam <orika@nvidia.com>
3 years agoregexdev: add resource limit reached flag
Ori Kam [Thu, 17 Dec 2020 10:37:30 +0000 (12:37 +0200)]
regexdev: add resource limit reached flag

When scanning a buffer it is possible that the scan will abort
due to some internal resource limit.

This commit adds such response flag, so application can handle such cases.

Signed-off-by: Francis Kelly <fkelly@nvidia.com>
Signed-off-by: Ori Kam <orika@nvidia.com>
3 years agoeal: add generic thread-local-storage functions
Tal Shnaiderman [Wed, 6 Jan 2021 20:35:53 +0000 (22:35 +0200)]
eal: add generic thread-local-storage functions

Add support for TLS functionality in EAL.

The following functions are added:
rte_thread_tls_key_create - create a TLS data key.
rte_thread_tls_key_delete - delete a TLS data key.
rte_thread_tls_value_set - set value bound to the TLS key
rte_thread_tls_value_get - get value bound to the TLS key

TLS key is defined by the new type rte_tls_key.

The API allocates the thread local storage (TLS) key.
Any thread of the process can subsequently use this key
to store and retrieve values that are local to the thread.

Those functions are added in addition to TLS capability
in rte_per_lcore.h to allow abstraction of the pthread
layer for all operating systems.

Windows implementation is under librte_eal/windows and
implemented using WIN32 API for Windows only.

Unix implementation is under librte_eal/unix and
implemented using pthread for UNIX compilation.

Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
3 years agoeal: move thread affinity functions to new file
Tal Shnaiderman [Wed, 6 Jan 2021 20:35:52 +0000 (22:35 +0200)]
eal: move thread affinity functions to new file

Move the definition of the functions
rte_thread_set_affinity and rte_thread_get_affinity
to new file, rte_thread.h

The file will implement generic threading functionality
and will only host threading functions which do not reference
pthread API.

Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
3 years agonet/i40e: refactor RSS flow
Alvin Zhang [Fri, 8 Jan 2021 05:35:40 +0000 (13:35 +0800)]
net/i40e: refactor RSS flow

1. Delete original code.
2. Add 2 tables(One maps flow pattern and RSS type to PCTYPE,
   another maps RSS type to input set).
3. Parse RSS pattern and RSS type to get PCTYPE.
4. Parse RSS action to get queues, RSS function and hash field.
5. Create and destroy RSS filters.
6. Create new files for hash flows.

Signed-off-by: Alvin Zhang <alvinx.zhang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/i40e: fix returned code for RSS hardware failure
Alvin Zhang [Fri, 8 Jan 2021 05:35:39 +0000 (13:35 +0800)]
net/i40e: fix returned code for RSS hardware failure

The API should return the system error status, but it returned the
hardware error status, this is confuses the caller.
This patch adds check on hardware execution status and returns -EIO
in case of hardware execution failure.

Fixes: 1d4b2b4966bb ("net/i40e: fix VF overwrite PF RSS LUT for X722")
Fixes: d0a349409bd7 ("i40e: support AQ based RSS config")
Cc: stable@dpdk.org
Signed-off-by: Alvin Zhang <alvinx.zhang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agodoc: fix RSS flow description in i40e guide
Alvin Zhang [Fri, 8 Jan 2021 05:35:38 +0000 (13:35 +0800)]
doc: fix RSS flow description in i40e guide

The command here does not create a queue region, but only sets the
lookup table, so the descriptions in the doc is not exact.

Fixes: feaae285b342 ("net/i40e: support hash configuration in RSS flow")
Cc: stable@dpdk.org
Signed-off-by: Alvin Zhang <alvinx.zhang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice/base: update copyright date
Qi Zhang [Fri, 8 Jan 2021 03:09:14 +0000 (11:09 +0800)]
net/ice/base: update copyright date

Updated the Copyright for 2021
Updated ice driver version.

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice/base: update add scheduler node counter
Qi Zhang [Fri, 8 Jan 2021 04:22:49 +0000 (12:22 +0800)]
net/ice/base: update add scheduler node counter

The number of nodes added counter was updated incorrectly. This issue
was exposed when the driver tried to add more than 128 queues per TC.

Fix added to update the counter correctly.

Fixes: 93e84b1bfc92 ("net/ice/base: add basic Tx scheduler")
Cc: stable@dpdk.org
Signed-off-by: Victor Raj <victor.raj@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice/base: cleanup style
Qi Zhang [Fri, 8 Jan 2021 04:20:26 +0000 (12:20 +0800)]
net/ice/base: cleanup style

A few style issues reported by checkpatch have snuck into the code,
resolve the style issues.

PARENTHESIS_ALIGNMENT: Alignment should match open parenthesis
COMPLEX_MACRO: Macros with complex values should be enclosed in parentheses

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice/base: support GTPU inner for AVF flow director
Qi Zhang [Fri, 8 Jan 2021 04:16:20 +0000 (12:16 +0800)]
net/ice/base: support GTPU inner for AVF flow director

Add dummy packets for IPV4_GTPU with inner IPV4/UDP/TCP with all
kinds of GTPU (EH) type (i.e., IP/EH/DL/UL) for AVF FDIR.

Signed-off-by: Junfeng Guo <junfeng.guo@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice/base: limit forced overrides based on FW version
Qi Zhang [Fri, 8 Jan 2021 04:10:55 +0000 (12:10 +0800)]
net/ice/base: limit forced overrides based on FW version

Beyond a specific version of firmware, there is no need to provide
override values to the firmware when setting PHY capabilities.  In this
case, we do not need to indicate whether we're in Strict or Lenient Link
Mode.

In the case of translating capabilities to the configuration structure,
the module compliance enforcement is already correctly set by firmware,
so the extra code block is redundant.

Signed-off-by: Jeb Cramer <jeb.j.cramer@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice/base: fix memory handling
Qi Zhang [Fri, 8 Jan 2021 04:03:52 +0000 (12:03 +0800)]
net/ice/base: fix memory handling

Fixed memory handling when memory allocated in user space was handled
as memory allocated in kernel space within QV os_dep implementation
of the ice_memdup function.

Fixes: 93e84b1bfc92 ("net/ice/base: add basic Tx scheduler")
Cc: stable@dpdk.org
Signed-off-by: Andrii Pypchenko <andrii.pypchenko@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice/base: add package ptype enable information
Qi Zhang [Fri, 8 Jan 2021 04:01:49 +0000 (12:01 +0800)]
net/ice/base: add package ptype enable information

Scan the 'Marker PType TCAM' session to retrieve the Rx parser PTYPE
enable information from the current package.

Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice/base: remove deprecated field
Qi Zhang [Fri, 8 Jan 2021 03:57:11 +0000 (11:57 +0800)]
net/ice/base: remove deprecated field

hw_vsi_id is used to replace vsi_id, so remove the deprecated vsi_id.

Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/ice/base: align add VSI and update VSI AQ command buffer
Qi Zhang [Fri, 8 Jan 2021 03:54:45 +0000 (11:54 +0800)]
net/ice/base: align add VSI and update VSI AQ command buffer

Aligned the buffer the following admin commands to their new
definitions:
* 0x210 = add_vsi
* 0x211 = update_vsi

Signed-off-by: Shay Amir <shay.amir@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
3 years agonet/virtio: improve logs in vhost-vDPA DMA mapping
Maxime Coquelin [Tue, 5 Jan 2021 15:34:46 +0000 (16:34 +0100)]
net/virtio: improve logs in vhost-vDPA DMA mapping

This patch adds debug logs in vhost_vdpa_dma_map() and
vhost_vdpa_dma_unmap() to ease debugging.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
3 years agovhost: refactor memory regions mapping
Maxime Coquelin [Tue, 5 Jan 2021 12:57:28 +0000 (13:57 +0100)]
vhost: refactor memory regions mapping

This patch moves memory region mmaping and related
preparation in a dedicated function in order to simplify
VHOST_USER_SET_MEM_TABLE request handling function.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
3 years agovhost: refactor postcopy registration
Maxime Coquelin [Tue, 5 Jan 2021 12:57:27 +0000 (13:57 +0100)]
vhost: refactor postcopy registration

This patch moves the registration of postcopy to a
dedicated function, with the goal of simplifying
VHOST_USER_SET_MEM_TABLE request handling function.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
3 years agovhost: refactor postcopy region registration
Maxime Coquelin [Tue, 5 Jan 2021 12:57:26 +0000 (13:57 +0100)]
vhost: refactor postcopy region registration

This patch moves the registration of memory regions to
userfaultfd to a dedicated function, with the goal of
simplifying VHOST_USER_SET_MEM_TABLE request handling
function.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
3 years agovdpa/mlx5: add hardware queue moderation
Xueming Li [Wed, 6 Jan 2021 03:06:30 +0000 (03:06 +0000)]
vdpa/mlx5: add hardware queue moderation

The next parameters control the HW queue moderation feature.
This feature helps to control the traffic performance and latency
trade-off.

Each packet completion report from HW to SW requires CQ processing by SW
and triggers interrupt for the guest driver. Interrupt report and
handling cost CPU cycles and time and the amount of this affects
directly on packet performance and latency.

hw_latency_mode parameters [int]
  0, HW default.
  1, Latency is counted from the first packet completion report.
  2, Latency is counted from the last packet completion.
hw_max_latency_us parameters [int]
  0 - 4095, The maximum time in microseconds that packet completion
  report can be delayed.
hw_max_pending_comp parameter [int]
  0 - 65535, The maximum number of pending packets completions in an HW
queue.

Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
3 years agocommon/mlx5: support vDPA completion queue moderation
Xueming Li [Wed, 6 Jan 2021 03:06:29 +0000 (03:06 +0000)]
common/mlx5: support vDPA completion queue moderation

This patch introduces new parameters for VirtQ CQ moderation, used for
performance tuning.

Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
3 years agovhost: replace SMP with thread fence for control path
Joyce Kong [Mon, 21 Dec 2020 15:50:33 +0000 (23:50 +0800)]
vhost: replace SMP with thread fence for control path

Simply replace the smp barriers with atomic thread fence for vhost control
path, if there are no synchronization points.

Signed-off-by: Joyce Kong <joyce.kong@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
3 years agovhost: replace SMP with thread fence for packed vring
Joyce Kong [Mon, 21 Dec 2020 15:50:32 +0000 (23:50 +0800)]
vhost: replace SMP with thread fence for packed vring

Simply replace smp barriers with atomic thread fence for
virtio packed vring.

Signed-off-by: Joyce Kong <joyce.kong@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
3 years agovhost: relax full barriers for used idx
Joyce Kong [Mon, 21 Dec 2020 15:50:31 +0000 (23:50 +0800)]
vhost: relax full barriers for used idx

Used idx can be synchronized by one-way barrier instead of full
write barrier for split vring.

Signed-off-by: Joyce Kong <joyce.kong@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
3 years agovhost: relax full barriers for desc flags
Joyce Kong [Mon, 21 Dec 2020 15:50:30 +0000 (23:50 +0800)]
vhost: relax full barriers for desc flags

Relax the full read barrier to one-way barrier for desc flags in
packed vring.

Signed-off-by: Joyce Kong <joyce.kong@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
3 years agovhost: remove unnecessary SMP barrier for avail idx
Joyce Kong [Mon, 21 Dec 2020 15:50:29 +0000 (23:50 +0800)]
vhost: remove unnecessary SMP barrier for avail idx

The ordering between avail index and desc reads has been enforced
by load-acquire for split vring, so smp_rmb barrier is not needed
behind it.

Signed-off-by: Joyce Kong <joyce.kong@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
3 years agovhost: remove unnecessary SMP barrier for desc flags
Joyce Kong [Mon, 21 Dec 2020 15:50:28 +0000 (23:50 +0800)]
vhost: remove unnecessary SMP barrier for desc flags

As function desc_is_avail performs a load-acquire barrier to
enforce the ordering between desc flags and desc content, it is
unnecessary to add a rte_smp_rmb barrier around the trace which
follows desc_is_avail.

Signed-off-by: Joyce Kong <joyce.kong@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>