aboutsummaryrefslogtreecommitdiffstats
path: root/resources/libraries
diff options
context:
space:
mode:
Diffstat (limited to 'resources/libraries')
-rw-r--r--resources/libraries/python/ContainerUtils.py82
-rw-r--r--resources/libraries/robot/shared/container.robot31
2 files changed, 96 insertions, 17 deletions
diff --git a/resources/libraries/python/ContainerUtils.py b/resources/libraries/python/ContainerUtils.py
index fc32248f6b..ef08317420 100644
--- a/resources/libraries/python/ContainerUtils.py
+++ b/resources/libraries/python/ContainerUtils.py
@@ -26,7 +26,8 @@ from resources.libraries.python.CpuUtils import CpuUtils
from resources.libraries.python.PapiExecutor import PapiSocketExecutor
from resources.libraries.python.ssh import SSH
from resources.libraries.python.topology import Topology, SocketType
-from resources.libraries.python.VppConfigGenerator import VppConfigGenerator
+from resources.libraries.python.VppConfigGenerator import (VppConfigGenerator,
+ VppInitConfig)
from resources.libraries.python.VPPUtil import VPPUtil
@@ -193,10 +194,7 @@ class ContainerManager:
def configure_vpp_in_all_containers(self, chain_topology, **kwargs):
"""Configure VPP in all containers.
- :param chain_topology: Topology used for chaining containers can be
- chain or cross_horiz. Chain topology is using 1 memif pair per
- container. Cross_horiz topology is using 1 memif and 1 physical
- interface in container (only single container can be configured).
+ :param chain_topology: Topology used for chaining containers.
:param kwargs: Named parameters.
:type chain_topology: str
:type kwargs: dict
@@ -220,47 +218,52 @@ class ContainerManager:
self.engine.container = self.containers[container]
guest_dir = self.engine.container.mnt[0].split(u":")[1]
- if chain_topology == u"chain":
+ if chain_topology == "chain":
self._configure_vpp_chain_l2xc(
mid1=mid1, mid2=mid2, sid1=sid1, sid2=sid2,
guest_dir=guest_dir, **kwargs
)
- elif chain_topology == u"cross_horiz":
+ elif chain_topology == "cross_horiz":
self._configure_vpp_cross_horiz(
mid1=mid1, mid2=mid2, sid1=sid1, sid2=sid2,
guest_dir=guest_dir, **kwargs
)
- elif chain_topology == u"chain_functional":
+ elif chain_topology == "chain_functional":
self._configure_vpp_chain_functional(
mid1=mid1, mid2=mid2, sid1=sid1, sid2=sid2,
guest_dir=guest_dir, **kwargs
)
- elif chain_topology == u"chain_ip4":
+ elif chain_topology == "chain_ip4":
self._configure_vpp_chain_ip4(
mid1=mid1, mid2=mid2, sid1=sid1, sid2=sid2,
guest_dir=guest_dir, **kwargs
)
- elif chain_topology == u"pipeline_ip4":
+ elif chain_topology == "pipeline_ip4":
self._configure_vpp_pipeline_ip4(
mid1=mid1, mid2=mid2, sid1=sid1, sid2=sid2,
guest_dir=guest_dir, **kwargs
)
- elif chain_topology == u"chain_vswitch":
+ elif chain_topology == "chain_vswitch":
self._configure_vpp_chain_vswitch(
mid1=mid1, mid2=mid2, sid1=sid1, sid2=sid2,
guest_dir=guest_dir, **kwargs)
- elif chain_topology == u"chain_ipsec":
+ elif chain_topology == "chain_ipsec":
idx_match = search(r"\d+$", self.engine.container.name)
if idx_match:
idx = int(idx_match.group())
self._configure_vpp_chain_ipsec(
mid1=mid1, mid2=mid2, sid1=sid1, sid2=sid2,
guest_dir=guest_dir, nf_instance=idx, **kwargs)
- elif chain_topology == u"chain_dma":
+ elif chain_topology == "chain_dma":
self._configure_vpp_chain_dma(
mid1=mid1, mid2=mid2, sid1=sid1, sid2=sid2,
guest_dir=guest_dir, **kwargs
)
+ elif chain_topology == "vswitch_ip4scale":
+ self._configure_vpp_vswitch_ip4scale(
+ mid1=mid1, mid2=mid2, sid1=sid1, sid2=sid2,
+ guest_dir=guest_dir, **kwargs
+ )
else:
raise RuntimeError(
f"Container topology {chain_topology} not implemented"
@@ -534,6 +537,47 @@ class ContainerManager:
vif1_mac=vif1_mac, vif2_mac=vif2_mac
)
+ def _configure_vpp_vswitch_ip4scale(self, **kwargs):
+ """Configure VPP in container with.
+
+ :param kwargs: Named parameters.
+ :type kwargs: dict
+ """
+ dut = self.engine.container.name.split("_")[0]
+ sid1 = kwargs["sid1"]
+ sid2 = kwargs["sid2"]
+ vid = kwargs["mid1"]
+
+ phy_cores = BuiltIn().get_variable_value("${cpu_count_int}")
+ rx_queues = BuiltIn().get_variable_value("${rxq_count_int}")
+ ifl = BuiltIn().get_variable_value("${int}")
+
+ if1_pci = Topology.get_interface_pci_addr(
+ self.engine.container.node,
+ BuiltIn().get_variable_value(f"${{{dut}_{ifl}{sid1}}}[0]")
+ )
+ if2_pci = Topology.get_interface_pci_addr(
+ self.engine.container.node,
+ BuiltIn().get_variable_value(f"${{{dut}_{ifl}{sid2}}}[0]")
+ )
+
+ compute_resource_info = CpuUtils.get_affinity_vswitch(
+ kwargs["nodes"], phy_cores, rx_queues=rx_queues
+ )
+ offset = (phy_cores + 1) * (vid - 1)
+ offset_cpu = compute_resource_info[f"{dut}_cpu_alloc_str"].split(",")
+ offset_cpu = [int(c) + int(offset) for c in offset_cpu]
+
+ print(offset_cpu)
+ self.engine.create_vpp_startup_config_vswitch(
+ offset_cpu, compute_resource_info["rxq_count_int"],
+ if1_pci, if2_pci
+ )
+
+ #self.engine.create_vpp_exec_config(
+ # "create_vswitch_ip4scale.exec"
+ #)
+
def stop_all_containers(self):
"""Stop all containers."""
# TODO: Rework if containers can be affected outside ContainerManager.
@@ -760,13 +804,17 @@ class ContainerEngine:
:type rxq: int
:type devices: list
"""
- vpp_config = self.create_base_vpp_startup_config(cpuset_cpus)
+ if cpuset_cpus is None:
+ cpuset_cpus = self.container.cpuset_cpus
+
+ vpp_config = VppInitConfig.create_vpp_startup_configuration_container(
+ self.container.node, cpuset_cpus
+ )
vpp_config.add_dpdk_dev(*devices)
- vpp_config.add_dpdk_log_level(u"debug")
+ vpp_config.add_dpdk_log_level("debug")
vpp_config.add_dpdk_no_tx_checksum_offload()
vpp_config.add_dpdk_dev_default_rxq(rxq)
- vpp_config.add_plugin(u"enable", u"dpdk_plugin.so")
- vpp_config.add_plugin(u"enable", u"perfmon_plugin.so")
+ vpp_config.add_plugin("enable", "dpdk_plugin.so")
# Apply configuration
self.execute(u"mkdir -p /etc/vpp/")
diff --git a/resources/libraries/robot/shared/container.robot b/resources/libraries/robot/shared/container.robot
index 9f08e6b1d4..dbcf690b72 100644
--- a/resources/libraries/robot/shared/container.robot
+++ b/resources/libraries/robot/shared/container.robot
@@ -371,3 +371,34 @@
| | END
| | Append To List | ${container_groups} | ${container_group}
| | Save VPP PIDs
+
+| Start vswitch containers
+| | [Documentation]
+| | ... | Configure and start multiple vswitch in container on all DUTs.
+| |
+| | ... | *Arguments:*
+| | ... | - phy_cores - Number of physical cores to use. Type: integer
+| | ... | - rx_queues: Number of RX queues. Type: integer
+| |
+| | [Arguments] | ${phy_cores} | ${rx_queues}=${None}
+| |
+| | Set Test Variable | @{container_groups} | @{EMPTY}
+| | Set Test Variable | ${container_group} | VSWITCH
+| | Import Library | resources.libraries.python.ContainerUtils.ContainerManager
+| | ... | engine=${container_engine} | WITH NAME | VSWITCH
+| | Stop VPP service on all DUTs | ${nodes}
+| | FOR | ${dut} | IN | @{duts}
+| | | FOR | ${i} | IN RANGE | 1 | ${${nic_pfs}//${2}+1}
+| | | | Construct container on DUT | ${dut}
+| | | | ... | nf_chains=${1} | nf_nodes=${${nic_pfs}//${2}}
+| | | | ... | nf_chain=${1} | nf_node=${i}
+| | | | ... | auto_scale=${False} | pinning=${False}
+| | | END
+| | END
+| | Run Keyword | VSWITCH.Acquire all containers
+| | Run Keyword | VSWITCH.Create all containers
+| | Run Keyword | VSWITCH.Configure vpp in all containers
+| | ... | vswitch_ip4scale | nodes=${nodes} | rts_per_flow=${rts_per_flow}
+| | Run Keyword | VSWITCH.Start VPP In All Containers
+| | Append To List | ${container_groups} | VSWITCH
+| | Save VPP PIDs