Current, the max waiting time for MBX response is 500ms, but in
some scenarios, it is not enough. Since it depends on the response
of the kernel mode driver, and its response time is related to the
scheduling of the system. In this special scenario, most of the
cores are isolated, and only a few cores are used for system
scheduling. When a large number of services are started, the
scheduling of the system will be very busy, and the reply of the
mbx message will time out, which will cause our PMD initialization
to fail.
This patch add a runtime config to set the max wait time. For the
above scenes, users can adjust the waiting time to a suitable value
by themselves.
Fixes: 463e748964 ("net/hns3: support mailbox")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
To support shared Rx queue, this patch introduces dedicate forwarding
engine. The engine groups received packets by mbuf->port into sub-group,
updates stream statistics and simply frees packets.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Shared Rx queue must be polled on same core. This patch checks and stops
forwarding if shared RxQ being scheduled on multiple
cores.
It's suggested to use same number of Rx queues and polling cores.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Xiaoyun Li <xiaoyun.li@intel.com>
In case of shared Rx queue, source port mbuf from polling result isn't
the Rx port of forwarding stream. To provide original port ID, this
patch dumps mbuf->port for each packet in verbose mode if shared Rx
queue enabled.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Xiaoyun Li <xiaoyun.li@intel.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Adds "--rxq-share=X" parameter to enable shared RxQ.
Rx queue is shared if device supports, otherwise fallback to standard
RxQ.
Shared Rx queues are grouped per X ports. X defaults to UINT32_MAX,
implies all ports join share group 1. Queue ID is mapped equally with
shared Rx queue ID.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Dump device capability and Rx domain ID if shared Rx queue is supported
by device.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Xiaoyun Li <xiaoyun.li@intel.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
This patch adds API to return name of device capability.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
In current DPDK framework, each Rx queue is pre-loaded with mbufs to
save incoming packets. For some PMDs, when number of representors scale
out in a switch domain, the memory consumption became significant.
Polling all ports also leads to high cache miss, high latency and low
throughput.
This patch introduces shared Rx queue. Ports in same Rx domain and
switch domain could share Rx queue set by specifying non-zero sharing
group in Rx queue configuration.
Shared Rx queue is identified by share_rxq field of Rx queue
configuration. Port A RxQ X can share RxQ with Port B RxQ Y by using
same shared Rx queue ID.
No special API is defined to receive packets from shared Rx queue.
Polling any member port of a shared Rx queue receives packets of that
queue for all member ports, port_id is identified by mbuf->port. PMD is
responsible to resolve shared Rx queue from device and queue data.
Shared Rx queue must be polled in same thread or core, polling a queue
ID of any member port is essentially same.
Multiple share groups are supported. PMD should support mixed
configuration by allowing multiple share groups and non-shared Rx queue
on one port.
Example grouping and polling model to reflect service priority:
Group1, 2 shared Rx queues per port: PF, rep0, rep1
Group2, 1 shared Rx queue per port: rep2, rep3, ... rep127
Core0: poll PF queue0
Core1: poll PF queue1
Core2: poll rep2 queue0
PMD advertise shared Rx queue capability via RTE_ETH_DEV_CAPA_RXQ_SHARE.
PMD is responsible for shared Rx queue consistency checks to avoid
member port's configuration contradict each other.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
In secondary process, rte_eth_dev_close() doesn't clear eth_dev->data.
If calling rte_dev_remove() after rte_eth_dev_close(), in
rte_eth_dev_pci_generic_remove() function, the released eth device still
can be found by its name in shared memory. As a result, the eth device
will be released repeatedly. The state of the eth device is modified to
RTE_ETH_DEV_UNUSED after rte_eth_dev_close(). So this state can be used
to avoid this problem.
Fixes: dcd5c8112b ("ethdev: add PCI driver helpers")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
This patch adds support for rte flow action type port_id to
enable directing packets from an input port PF to an output
port which is a VF of the input port PF.
Signed-off-by: Satheesh Paul <psatheesh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
This patch adds ROC API to support flow port ID action type.
Signed-off-by: Satheesh Paul <psatheesh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Vhost will update desc’s Buffer ID advance to next used descriptor when
VIRTIO_F_IN_ORDER feature negotiated. When virtio reuses the descriptor,
the Buffer ID should be restored even VIRTQ_DESC_F_INDIRECT
feature negotiated.
Fixes: b473061b0e ("net/virtio: fix indirect descriptors in packed datapaths")
Cc: stable@dpdk.org
Signed-off-by: Xuan Ding <xuan.ding@intel.com>
Signed-off-by: Yong Liu <yong.liu@intel.com>
Signed-off-by: Miao Li <miao.li@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
To improve performance in vhost Tx/Rx, merge vhost stats loop.
eth_vhost_tx has 2 loop of send num iteraion.
It can be merge into one.
eth_vhost_rx has the same issue as Tx.
Signed-off-by: Gaoxiang Liu <liugaoxiang@huawei.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
The use of IOMMU has many advantages, such as isolation and address
translation. This patch extends the capability of DMA engine to use
IOMMU if the DMA engine is bound to vfio.
When set memory table, the guest memory will be mapped
into the default container of DPDK.
Signed-off-by: Xuan Ding <xuan.ding@intel.com>
Tested-by: Yvonne Yang <yvonnex.yang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Currently, if we map a memory area A, then map a separate memory area B
that by coincidence happens to be adjacent to A, current implementation
will merge these two segments into one, and if partial unmapping is not
supported, these segments will then be only allowed to be unmapped in
one go. In other words, given segments A and B that are adjacent, it
is currently not possible to map A, then map B, then unmap A.
Fix this by adding a notion of "chunk size", which will allow
subdividing segments into equally sized segments whenever we are dealing
with an IOMMU that does not support partial unmapping. With this change,
we will still be able to merge adjacent segments, but only if they are
of the same size. If we keep with our above example, adjacent segments A
and B will be stored as separate segments if they are of different
sizes.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Signed-off-by: Xuan Ding <xuan.ding@intel.com>
Tested-by: Yvonne Yang <yvonnex.yang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Add initialization for packed ring indirect descriptors
in reconnection path.
Fixes: 381f39ebb7 ("net/virtio: fix packed ring indirect descricptors setup")
Cc: stable@dpdk.org
Signed-off-by: Xuan Ding <xuan.ding@intel.com>
Tested-by: Yinan Wang <yinan.wang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
The index in rte_vhost_set_last_inflight_io_split is from
the frontend driver, check if it's in the virtqueue range.
Fixes: bb0c2de960 ("vhost: add APIs to operate inflight ring")
Cc: stable@dpdk.org
Signed-off-by: Li Feng <fengli@smartx.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Tx prepare method calls rte_net_intel_cksum_prepare(), which
handles tunnel packets correctly, but Tx burst path does not
take tunnel presence into account when computing the offsets.
Fixes: 58169a9c81 ("net/virtio: support Tx checksum offload")
Cc: stable@dpdk.org
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
Reviewed-by: Olivier Matz <olivier.matz@6wind.com>
When a vdev is removed in destroy_device function,
the corresponding vhost TX buffer will also be freed,
but the vhost TX buffer may still be used in the
drain_vhost function, which will cause an error of
heap-use-after-free. Therefore, before accessing
vhost TX buffer, we need to check whether the vdev
has been removed, if so, let's skip this vdev.
Fixes: a68ba8e0a6 ("examples/vhost: refactor vhost data path")
Cc: stable@dpdk.org
Signed-off-by: Wenwu Ma <wenwux.ma@intel.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
If packed ring size is not power of two, it is possible that remained
number less than one batch and meanwhile batch operation can pass.
This will cause incorrect remained number calculation and then lead to
receiving oversized packets. The patch fixed the issue by added
remained number check before batch operation.
Fixes: 77d66da838 ("net/virtio: add vectorized packed ring Rx")
Cc: stable@dpdk.org
Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
VAR is the device memory space for the virtio queues doorbells,
Qemu could mmap it to directly to speed up doorbell push.
On a busy system, Qemu takes time to release VAR resources during driver
shutdown. If vdpa restarted quickly, the VAR allocation failed with
error 28 since the VAR is singleton resource per device.
This patch adds retry mechanism for VAR allocation.
Fixes: 4cae722c1b ("vdpa/mlx5: move virtual doorbell alloc to probe")
Cc: stable@dpdk.org
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Matan Azrad <matan@nvidia.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
After a vDPA application restart, Qemu restores VQ with used and
available index, new incoming packet triggers virtio driver to
handle buffers. Under heavy traffic, no available buffer for
firmware to receive new packets, no Rx interrupts generated,
driver is stuck on endless interrupt waiting.
As a firmware workaround, this patch sends a notification after
VQ setup to ask driver handling buffers and filling new buffers.
Fixes: bff7350110 ("vdpa/mlx5: prepare virtio queues")
Cc: stable@dpdk.org
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Matan Azrad <matan@nvidia.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
This patch fixes the wrong way to obtain virtqueue.
The end of virtqueue cannot be judged based on whether
the array is NULL.
Fixes: 4e8169eb0d ("net/virtio: fix Rx scatter offload")
Cc: stable@dpdk.org
Signed-off-by: Zhihong Peng <zhihongx.peng@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
After DCF commits TM hierarchy configuration, the commit flag is set to
avoid duplicated commit. But the flag is not reset after device stop,
which prevents the update of hierarchy configuration unless close the
device. It is not reasonable. This patch fix to reset the commit flag
after device stop. Then users can delete and add nodes to commit a new
TM hierarchy configuration.
Fixes: 3a6bfc37ea ("net/ice: support QoS config VF bandwidth in DCF")
Cc: stable@dpdk.org
Signed-off-by: Ting Xu <ting.xu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
This patch enables building the e1000 driver for Windows.
I tested using two Windows VM on top of VMware Fusion,
creating two e1000 devices with device ID 0x10D3 (8274L),
verifying rx/tx works correctly using dpdk-testpmd.exe
rxonly and txonly mode.
Signed-off-by: William Tu <u9012063@gmail.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Pallavi Kadam <pallavi.kadam@intel.com>
Tested-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
Tested-by: Pallavi Kadam <pallavi.kadam@intel.com>
On a VMware ESXi 6.0 setup with an Intel 82599 NIC the ports don't
seem to initialize anymore, while running testpmd.
Configuring Port 0 (socket 0)
ixgbevf_dev_rx_init(): Set max packet length to 1518 failed.
ixgbevf_dev_start(): Unable to initialize RX hardware (-22)
Fail to start port 0: Invalid argument
Configuring Port 1 (socket 0)
ixgbevf_dev_rx_init(): Set max packet length to 1518 failed.
ixgbevf_dev_start(): Unable to initialize RX hardware (-22)
Fail to start port 1: Invalid argument
Please stop the ports first
If the call to ixgbevf_rlpml_set_vf fails and we return prematurely,
we will not be able to initialize the ports correctly.
The behavior seems to have changed since the following commit:
Fixes: c77866a169 ("net/ixgbe: detect failed VF MTU set")
Cc: stable@dpdk.org
We can make this particular use case work correctly if we don't
return an error, which seems to be consistent with the overall
kernel ixgbevf implementation.
[1]
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/
drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c?h=v5.14#n2015
Signed-off-by: Tudor Cornea <tudor.cornea@gmail.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
Previously, we set txq affinity to 0 and let firmware
to perform round-robin when bonding. Firmware uses a
global counter to assign txq affinity to different
physical ports accord to remainder after division.
There are three dis-advantages:
1. The global counter is shared between kernel and dpdk.
2. After restarting pmd or port, the previous counter value
is reused, so the new affinity is unpredictable.
3. There is no way to get what affinity is set by firmware.
In this update, we will create several TISs up to the
number of bonding ports and bind each TIS to one PF port.
For each port, it will start to pick up TIS using its port
index. Upper layer application can quickly calculate each txq's
affinity without querying.
At DPDK layer, when creating txq with 2 bonding ports, the
affinity is set like:
port 0: 1-->2-->1-->2
port 1: 2-->1-->2-->1
port 2: 1-->2-->1-->2
Note: Only applicable to DevX api.
This affinity subjects to HW hash.
Signed-off-by: Rongwei Liu <rongweil@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Added a new function mlx5_devx_cmd_query_lag() to query LAG
property from firmware including state/affinity/mode etc.
Signed-off-by: Jiawei Wang <jiaweiw@nvidia.com>
Signed-off-by: Rongwei Liu <rongweil@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
MLX5 PMD exposes a socket for external tools to dump port state.
Socket events are listened using an interrupt source of EXT type.
The socket was closed and the interrupt callback was unregistered
at program exit, which is incorrect because DPDK could be already
shut down at this point. Move actions performed at program exit
to the moment the last MLX5 port is closed. The socket will be opened
again if later a new MLX5 device is plugged in and probed.
Also fix comments that were decisively talking
about secondary processes instead of external tools.
Fixes: e6cdc54cc0 ("net/mlx5: add socket server for external tools")
Cc: stable@dpdk.org
Reported-by: Harman Kalra <hkalra@marvell.com>
Signed-off-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
mlx5_rxq_start() allocates rxq_ctrl->obj and frees it on failure,
but did not set it to NULL. Later mlx5_rxq_release() could not recognize
this object is already freed and attempted to release its resources,
resulting in a crash:
Configuring Port 0 (socket 0)
mlx5_common: Failed to create RQ using DevX
mlx5_common: Can't create DevX RQ object.
mlx5_net: Port 0 Rx queue 0 RQ creation failure.
Segmentation fault
Set rxq_ctrl->obj to NULL after it is freed to skip resource release.
Fixes: 1260a87b28 ("net/mlx5: share Rx control code")
Cc: stable@dpdk.org
Signed-off-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
The RSS configuration in a policy action container was a pointer
inside a union, and the pointer area could be used as other fate
action. In the current implementation, the RSS of the green color
was prior to that of the yellow color. There was a high possibility
the pointer was considered as the RSS and result in a error flow
expansion when only the yellow color had the RSS action.
The check of the fate action type should also be done to get rid of
the misjudgment.
Fixes: b38a12272b ("net/mlx5: split meter color policy handling")
Cc: stable@dpdk.org
Signed-off-by: Bing Zhao <bingz@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Verbs API doesn't support device port number larger than 255 by design.
To support more VF or SubFunction port representors, forces DevX API
check when max Verbs device link ports larger than 255.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Verbs API does not support Infiniband device port number larger 255 by
design. To support more representors on a single Infiniband device DevX
API should be engaged.
While creating Send Queue (SQ) object with Verbs API, the PMD assigned
IB device port attribute and kernel created the default miss flows in
FDB domain, to redirect egress traffic from the queue being created to
representor appropriate peer (wire, HPF, VF or SF).
With DevX API there is no IB-device port attribute (it is merely kernel
one, DevX operates in PRM terms) and PMD must create default miss flows
in FDB explicitly. PMD did not provide this and using DevX API for
E-Switch configurations was disabled.
The default miss FDB flow matches E-Switch manager vport (to make sure
the source is some representor) and SQn (Send Queue number - device
internal queue index). The root flow table managed by kernel/firmware
and it does not support vport redirect action, we have to split the
default miss flow into two ones:
- flow with lowest priority in the root table that matches E-Switch
manager vport ID and jump to group 1.
- flow in group 1 that matches E-Switch manager vport ID and SQn and
forwards packet to peer vport
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
When creating internal transfer flow on root table with lowest
priority, the flow was created with max UINT32_MAX priority. It is wrong
since the flow is created in kernel and max priority supported is 16.
This patch fixes this by adding internal flow check.
Fixes: 5f8ae44dd4 ("net/mlx5: enlarge maximal flow priority")
Cc: stable@dpdk.org
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Extends txq flow pattern to support both hairpin and regular txq.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
For egress packet on representor, the vport ID in transport domain
is E-Switch manager vport ID since representor shares resources of
E-Switch manager. E-Switch manager vport ID and Tx queue internal device
index are used to match representor egress packet.
This patch adds flow item port ID match on E-Switch manager.
E-Switch manager vport ID is 0xfffe on BlueField, 0 otherwise.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
To detect number flow Verbs flow priorities, PMD try to create Verbs
flows in different priority. While Verbs is not designed to support
ports larger than 255.
When DevX supported by kernel driver, 16 Verbs priorities must be
supported, no need to create Verbs flows.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
IB spec doesn't allow 255 ports on a single HCA, port number of 256 was
cast to u8 value 0 which invalid to ibv_query_port()
This patch invokes Netlink API to query port state when port number
greater than 255.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Introduce netlink API to get RDMA port state.
Port state is retrieved based on RDMA device name and port index.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Add support for test-pmd to parse protocol pattern L2TPv2 and PPP.
Signed-off-by: Wenjun Wu <wenjun1.wu@intel.com>
Signed-off-by: Jie Wang <jie1x.wang@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Add support for PPP over L2TPv2 over UDP protocol RSS Hash based
on inner IP src/dst address and TCP/UDP src/dst port.
Patterns are listed below:
eth/ipv4(6)/udp/l2tpv2/ppp/ipv4(6)
eth/ipv4(6)/udp/l2tpv2/ppp/ipv4(6)/udp
eth/ipv4(6)/udp/l2tpv2/ppp/ipv4(6)/tcp
Signed-off-by: Wenjun Wu <wenjun1.wu@intel.com>
Signed-off-by: Jie Wang <jie1x.wang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Added flow pattern items and header formats of L2TPv2 and PPP.
Signed-off-by: Wenjun Wu <wenjun1.wu@intel.com>
Signed-off-by: Jie Wang <jie1x.wang@intel.com>
Acked-by: Ori Kam <orika@nvidia.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
CNF10KA does not differ it terms of RVU resources from
CN10KA platform hence add it to list of devices respective
drivers support.
Otherwise devices on CNF10KA are not probed even though
compatible drivers exist.
Signed-off-by: Tomasz Duszynski <tduszynski@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Full stop at the end of short comment just make line longer. It
should be either everywhere or nowhere to be consistent.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Add empty lines to separate fields commented using different styles.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ori Kam <orika@nvidia.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ori Kam <orika@nvidia.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
