Difference between revisions of "VPP/Command-line Arguments"

From fd.io
< VPP
Jump to: navigation, search
(experimentally tabularize this section)
(tabularize the rest of the page)
Line 66: Line 66:
 
The "tuntap" driver configures a point-to-point interface between the vpp engine and the local Linux kernel stack. This allows e.g. users to ssh to the host / VM / container via vpp "revenue" interfaces. It's marginally useful, and is currently disabled by default. To [dynamically] create TAP interfaces - the preferred scheme - see the "tap_connect" binary API  
 
The "tuntap" driver configures a point-to-point interface between the vpp engine and the local Linux kernel stack. This allows e.g. users to ssh to the host / VM / container via vpp "revenue" interfaces. It's marginally useful, and is currently disabled by default. To [dynamically] create TAP interfaces - the preferred scheme - see the "tap_connect" binary API  
  
*"tuntap disable" -- disable the tun/tap driver if desired  
+
{| class="wikitable"
*"tuntap ethernet" -- create a tap device (ethernet MAC) instead of a tun device (point-to-point tunnel)  
+
| <code>tuntap disable</code>
*"tuntap have-normal-interface" -- treat the host Linux stack as a routing peer instead of programming vpp interface L3 addresses onto the tun/tap devices. You'll need to manually configure the Linux network stack "vnet" interface, configure vlan subinterfaces if desired, etc.  
+
| Disable the tun/tap driver if desired  
*"tuntap name ishmael" -- call the tun/tap device ishmael
+
|-
 +
| <code>tuntap ethernet</code>
 +
| Create a tap device (ethernet MAC) instead of a tun device (point-to-point tunnel)  
 +
|-
 +
| <code>tuntap have-normal-interface</code>
 +
| Treat the host Linux stack as a routing peer instead of programming vpp interface L3 addresses onto the tun/tap devices. You'll need to manually configure the Linux network stack "vnet" interface, configure vlan subinterfaces if desired, etc.  
 +
|-
 +
| <code>tuntap name ishmael</code>
 +
| Call the tun/tap device ishmael
 +
|}
  
 
Here's a typical multiple parameter invocation:  
 
Here's a typical multiple parameter invocation:  
Line 79: Line 88:
 
The ability to trace, dump, and replay control-plane API traces makes all the difference in the world when trying to understand what the control-plane has tried to ask the forwarding-plane to do.  
 
The ability to trace, dump, and replay control-plane API traces makes all the difference in the world when trying to understand what the control-plane has tried to ask the forwarding-plane to do.  
  
*"on" -- enable API trace capture from the beginning of time, and arrange for a post-mortem dump of the API trace if the application terminates abnormally. By default, the (circular) trace buffer will be configured to capture 256K traces
+
{| class="wikitable"
 
+
| <code>on</code>
*"nitems &lt;nnn&gt;" -- Configure the circular trace buffer to contain the last &lt;nnn&gt; entries. By default, the trace buffer captures the last 256K API messages received.
+
| Enable API trace capture from the beginning of time, and arrange for a post-mortem dump of the API trace if the application terminates abnormally. By default, the (circular) trace buffer will be configured to capture 256K traces
 +
|-
 +
| <code>nitems &lt;nnn&gt;</code>
 +
| Configure the circular trace buffer to contain the last &lt;nnn&gt; entries. By default, the trace buffer captures the last 256K API messages received.
 +
|}
  
 
Typically, one simply enables the API message trace scheme:  
 
Typically, one simply enables the API message trace scheme:  
Line 91: Line 104:
 
Command-line CPU configuration controls the creation of named thread types, and the cpu affinity thereof. As of this writing, the cpu configuration function supports these parameters:  
 
Command-line CPU configuration controls the creation of named thread types, and the cpu affinity thereof. As of this writing, the cpu configuration function supports these parameters:  
  
*"workers &lt;nn&gt;" - create &lt;nn&gt; worker threads.  
+
{| class="wikitable"
*"io &lt;nn&gt;" - create &lt;nn&gt; i/o threads.  
+
| <code>workers &lt;nn&gt;</code>
*"main-thread-io" - handle i/o devices from thread 0, hand off traffic to worker threads. Requires "workers &lt;nn&gt;".  
+
| Create &lt;nn&gt; worker threads.
*"skip-cores &lt;nn&gt;" - leave the low &lt;nn&gt; bits of the process affinity mask clear.  
+
|-
 +
| <code>io &lt;nn&gt;</code>
 +
| Create <code>nn</code> i/o threads.
 +
|-
 +
| <code>main-thread-io</code>
 +
| Handle i/o devices from thread 0, hand off traffic to worker threads. Requires "workers &lt;nn&gt;".  
 +
|-
 +
| <code>skip-cores &lt;nn&gt;</code>
 +
| Leave the low <code>nn</code> bits of the process affinity mask clear.  
 +
|}
 +
 
 
In some cases, it may be necessary to place thread instances manually:  
 
In some cases, it may be necessary to place thread instances manually:  
  
*"main-core &lt;n&gt;" - assign main thread to specific core&nbsp;
+
{| class="wikitable"
*"coremask-workers &lt;hex-mask&gt;" - place worker threads according to &lt;hex-mask&gt;
+
| <code>main-core &lt;n&gt;</code>
*"corelist-workers &lt;list&gt;" - same like coremask-workers but accepts list of cores instead of bitmap
+
| Assign main thread to specific core.
*"coremask-io &lt;hex-mask&gt;" - place i/o threads according to &lt;hex-mask&gt;
+
|-
*"corelist-io &lt;list&gt;" - same like coremask-io but accepts list of cores instead of bitmap
+
| <code>coremask-workers &lt;hex-mask&gt;</code>
 +
| Place worker threads according to the bitmap <code>hex-mask</code>.
 +
|-
 +
| <code>corelist-workers &lt;list&gt;</code>
 +
| Same as <code>coremask-workers</code> but accepts a list of cores instead of a bitmap.
 +
|-
 +
| <code>coremask-io &lt;hex-mask&gt;</code>
 +
| Place I/O threads according to the bitmap <code>hex-mask</code>.
 +
|-
 +
| <code>corelist-io &lt;list&gt;</code>
 +
| Same as <code>coremask-io</code> but accepts a list of cores instead of a bitmap.
 +
|}
  
Note that the "main" thread always occupies the lowest core-id specified in the dpdk [process-level] coremask.  
+
Note that the "main" thread always occupies the lowest core-id specified in the DPDK [process-level] coremask.  
  
 
Here's a full-bore manual placement example:  
 
Here's a full-bore manual placement example:  
Line 120: Line 154:
 
Command line DPDK configuration controls a number of parameters, including device whitelisting, the number of CPUs available for launching dpdk-eal-controlled threads, the number of I/O buffers, and the process affinity mask.  
 
Command line DPDK configuration controls a number of parameters, including device whitelisting, the number of CPUs available for launching dpdk-eal-controlled threads, the number of I/O buffers, and the process affinity mask.  
  
In addition, the dpdk configuration function attempts to support all of the dpdk eal configuration parameters.  
+
In addition, the DPDK configuration function attempts to support all of the DPDK EAL configuration parameters.  
  
 
Popular options include:  
 
Popular options include:  
  
*coremask &lt;hex&gt; - sets the process-level coremask. See [[VPP/Command-line_Arguments#.22cpu.22_parameters|previous section on cpu parameters]] for more details.  
+
{| class="wikitable"
*no-hugetlb - don't use huge TLB pages. Potentially useful for running simulator images.  
+
| <code>coremask &lt;hex&gt;</code>
*dev &lt;pci-dev&gt; - white-list [as in, attempt to drive] a specific PCI device. PCI-dev is a string of the form "DDDD:BB:SS.F" where DDDD = Domain, BB = Bus Number, SS = Slot number, and F = Function. This is the same format used in the linux sysfs tree (i.e. /sys/bus/pci/devices) for PCI device directory names.  
+
| Sets the process-level coremask. See [[VPP/Command-line_Arguments#.22cpu.22_parameters|previous section on cpu parameters]] for more details.  
*vdev &lt;DPDK EAL command to create link bonding interfaces&gt; - specify bonded Ethernet interfaces, operating modes and PCI address of slave links. For example, "vdev eth_bond0,mode=2,slave=0000:0f:00.0,slave=0000:11:00.0,xmit_policy=l34 vdev eth_bond1,mode=2,slave=0000:10:00.0,slave=0000:12:00.0,xmit_policy=l34". Only XOR balanced (mode 2) mode is supported.  
+
|-
*num-mbufs &lt;nn&gt; - number of I/O buffers. You need to increase this if you have a lot of physical interfaces with multiple RSS queues (for example, 131072).
+
| <code>no-hugetlb</code>
*kni &lt;nn&gt; - number of KNI interfaces. Refer to the dpdk documentation  
+
| Don't use huge TLB pages. Potentially useful for running simulator images.  
*uio-driver &lt;name&gt; - defaults to "igb_uio" [the Ubuntu default]. The names "uio_pci_generic" or "vfio-pci" are other possible values.
+
|-
 +
| <code>dev &lt;pci-dev&gt;</code>
 +
| White-list [as in, attempt to drive] a specific PCI device. PCI-dev is a string of the form "<code>DDDD:BB:SS.F</code>" where <code>DDDD</code> = Domain, <code>BB</code> = Bus Number, <code>SS</code> = Slot number, and <code>F</code> = Function. This is the same format used in the linux sysfs tree (i.e. <code>/sys/bus/pci/devices</code>) for PCI device directory names.  
 +
|-
 +
| <code>vdev &lt;DPDK EAL command to create link bonding interfaces&gt;</code>
 +
| Specify bonded Ethernet interfaces, operating modes and PCI address of slave links. For example, "<code>vdev eth_bond0,mode=2,slave=0000:0f:00.0,slave=0000:11:00.0,xmit_policy=l34 vdev eth_bond1,mode=2,slave=0000:10:00.0,slave=0000:12:00.0,xmit_policy=l34</code>". Only XOR balanced (mode 2) mode is supported.  
 +
|-
 +
| <code>num-mbufs &lt;nn&gt;</code>
 +
| Number of I/O buffers. You need to increase this if you have a lot of physical interfaces with multiple RSS queues (for example, <code>131072</code>).
 +
|-
 +
| <code>kni &lt;nn&gt;</code>
 +
| Number of KNI interfaces. Refer to the DPDK documentation  
 +
|-
 +
| <code>uio-driver &lt;name&gt;</code>
 +
| Defaults to "<code>igb_uio</code>" [the Ubuntu default]. The names "<code>uio_pci_generic</code>" or "<code>vfio-pci</code>" are other possible values.
 +
|}
  
All of the DPDK EAL options should be available. See .../open-repo/vnet/vnet/devices/dpdk/dpdk.h, look at the set of "foreach_eal_XXX" macros.
+
All of the DPDK EAL options should be available. See <code>.../open-repo/vnet/vnet/devices/dpdk/dpdk.h</code>, look at the set of "<code>foreach_eal_XXX</code>" macros.
  
 
==== "l2tp" parameters  ====
 
==== "l2tp" parameters  ====
Line 138: Line 187:
 
ip6-l2tpv3 configuration controls the method used to locate a specific ip6-l2tpv3 tunnel. The following settings are mutually exclusive  
 
ip6-l2tpv3 configuration controls the method used to locate a specific ip6-l2tpv3 tunnel. The following settings are mutually exclusive  
  
*lookup-v6-src - lookup tunnel by ip6 source address  
+
{| class="wikitable"
*lookup-v6-dst - lookup tunnel by ip6 destination address  
+
| <code>lookup-v6-src</code>
*lookup-session-id - lookup tunnel by l2tpv3 session identifier
+
| Lookup tunnel by ip6 source address.
 +
|-
 +
| <code>lookup-v6-dst</code>
 +
| Lookup tunnel by ip6 destination address.
 +
|-
 +
| <code>lookup-session-id</code>
 +
| Lookup tunnel by l2tpv3 session identifier.
 +
|}
  
Refer to .../open-repo/vnet/vnet/devices/dpdk/{dpdk.h, init.c} for more details  
+
Refer to <code>.../open-repo/vnet/vnet/devices/dpdk/{dpdk.h, init.c}</code> for more details.
  
 
==== "oam" parameters  ====
 
==== "oam" parameters  ====
  
Oam configuration controls the (ip4-icmp) interval, and number of misses allowed before reporting an oam target down to any registered listener  
+
Oam configuration controls the (ip4-icmp) interval, and number of misses allowed before reporting an oam target down to any registered listener.
  
*interval &lt;floating-point-seconds&gt; - interval between sending oam ip4 icmps
+
{| class="wikitable"
*misses-allowed &lt;nn&gt; - number of misses before declaring an oam target down
+
| <code>interval &lt;floating-point-seconds&gt;</code>
 +
| Interval between sending oam ip4 icmp messages.
 +
|-
 +
| <code>misses-allowed &lt;nn&gt;</code>
 +
| Number of misses before declaring an oam target down.
 +
|}
  
 
==== "heapsize" parameter  ====
 
==== "heapsize" parameter  ====
Line 155: Line 216:
 
Heapsize configuration controls the size of the main heap. The heap size is configured very early in the boot sequence, before loading plug-ins or doing much of anything else.  
 
Heapsize configuration controls the size of the main heap. The heap size is configured very early in the boot sequence, before loading plug-ins or doing much of anything else.  
  
*heapsize &lt;nn&gt;M | &lt;nn&gt;G - specifies the size of the heap in mb or gb. Setting the main heap size to 4gb or more requires recompilation of the entire system with CLIB_VEC64 &gt; 0. See .../clib/clib/vec_bootstrap.h.
+
{| class="wikitable"
 +
| <code>heapsize &lt;nn&gt;M | &lt;nn&gt;G</code>
 +
| Specifies the size of the heap in mb or gb. Setting the main heap size to 4gb or more requires recompilation of the entire system with <code>CLIB_VEC64 &gt; 0</code>. See <code>.../clib/clib/vec_bootstrap.h</code>.
 +
|}
  
 
==== "plugin_path" parameter  ====
 
==== "plugin_path" parameter  ====
Line 161: Line 225:
 
Plugin_path configuration controls the set of directories searched for vlib plugins. Supply a colon-separated list of (absolute) directory names:  
 
Plugin_path configuration controls the set of directories searched for vlib plugins. Supply a colon-separated list of (absolute) directory names:  
  
*plugin_path dir1:dir2:...:dirN
+
{| class="wikitable"
 +
| <code>plugin_path dir1:dir2:...:dirN</code>
 +
| Configure the search path for plugins.
 +
|}
  
 
==== "cj" parameters  ====
 
==== "cj" parameters  ====
  
The cj thread-safe circular log buffer scheme is occasionally useful when chasing bugs. Calls to it should not be checked in. See .../vlib/vlib/unix/cj.c.  
+
The cj thread-safe circular log buffer scheme is occasionally useful when chasing bugs. Calls to it should not be checked in. See <code>.../vlib/vlib/unix/cj.c</code>.
  
*records &lt;nn&gt; - configure the number of cj records in the circular buffer  
+
{| class="wikitable"
*on - turns on logging at the earliest possible moment
+
| <code>records &lt;nn&gt;</code>
 +
| Configure the number of cj records in the circular buffer.
 +
|-
 +
| <code>on</code>
 +
| Turns on logging at the earliest possible moment.
 +
|}
  
 
==== "vhost-user" parameters  ====
 
==== "vhost-user" parameters  ====
Line 174: Line 246:
 
Vhost-user configuration parameters control the vhost-user driver.  
 
Vhost-user configuration parameters control the vhost-user driver.  
  
*coalesce-frames &lt;nn&gt; - Subject to deadline-timer expiration - see next item - attempt to transmit at least NN packet frames.  
+
{| class="wikitable"
*coalesce-time &lt;fp-seconds&gt; - hold packets no longer than &lt;floating-point-seconds&gt; before transmitting them  
+
| <code>coalesce-frames &lt;nn&gt;</code>
*dont-dump-memory - vhost-user shared-memory segments can add up to a large amount of memory, so it's handy to avoid adding them to corefiles when using a significant number of such interfaces.
+
| Subject to deadline-timer expiration - see next item - attempt to transmit at least <code>nn</code> packet frames.  
 +
|-
 +
| <code>coalesce-time &lt;seconds&gt;</code>
 +
| Hold packets no longer than (floating-point) <code>seconds</coode> before transmitting them.
 +
|-
 +
| <code>dont-dump-memory</code>
 +
| vhost-user shared-memory segments can add up to a large amount of memory, so it's handy to avoid adding them to corefiles when using a significant number of such interfaces.  
 +
|}
  
 
==== "l2learn" parameter  ====
 
==== "l2learn" parameter  ====
Line 182: Line 261:
 
Configures the number of L2 (MAC) addresses in the L2 FIB at any one time. Defaults to 1M entries.  
 
Configures the number of L2 (MAC) addresses in the L2 FIB at any one time. Defaults to 1M entries.  
  
limit &lt;nn&gt; - limits the size of the L2 FIB to &lt;nn&gt; concurrent entries.
+
{| class="wikitable"
 +
| <code>limit &lt;nn&gt;</code>
 +
| Limits the size of the L2 FIB to <code>nn</code> concurrent entries.
 +
|}

Revision as of 03:02, 28 April 2016

Introduction

The vpp network stack comes with a command-line interface that can be used for debugging purposes.

Please help maintain and improve this document!

Command-line arguments

The vpp stack can be configured from the command-line in a number of ways. Specific applications built on the stack have been known to require a dozen arguments, depending on requirements. This section describes commonly-used command-line options. These options are parsed by individual bits of code, similar to the action functions of the CLI parser.

You can find command-line argument parsers in the source code by searching for instances of the VLIB_CONFIG_FUNCTION macro. The actual command-line invocation syntax is extremely flexible.

macro invocation: VLIB_CONFIG_FUNCTION (foo_config, "foo")
command-line: "foo { arg1 arg2 arg3 ... }"

When using the command on the command line, surround the list of arguments with curly braces.

NOTE: If you supply a single argument to a specific configuration function, you do not need to surround the argument with curly braces. For example, the following command argument does not have braces:

# /cisco/bin/swt unix interactive
SWT: <start-typing..>

Vpp applications must be able to locate their own executable images. The simplest way to ensure this will work: invoke a vpp application by giving its absolute path. At startup, vpp aplications parse through their own Elf-sections [primarily] to make lists of init, configuration, and exit handlers. In gdb, it's often sufficient to start an application like this:

(gdb) r unix interactive

"unix" parameters

interactive Attach CLI to stdin/out and provide a debugging command line interface. Implies nodaemon.
nodaemon Do not fork / background the vpp process. Typical when invoking VPP applications from a process monitor.
cli-listen localhost:5002 Bind the CLI to listen at address localhost on TCP port 5002.
exec <filename> Read startup configuration from filename.
log <filename> Logs the startup configuration and all subsequent CLI commands in filename. Very useful in situations where folks don't remember or can't be bothered to include CLI commands in bug reports.
cli-line-mode Disable character-by-character I/O on stdin. Useful when combined with, for example, emacs M-x gud-gdb.
cli-history-limit <nn> Limit commmand history to nn lines. A value of 0 disables command history. Default value: 50
cli-no-banner Disable the login banner on stdin and Telnet connections.
cli-no-pager Disable the output pager.
cli-pager-buffer-limit <nn> Limit pager buffer to nn lines of output. A value of 0 disables the pager. Default value: 100000
full-coredump Ask the Linux kernel to dump all memory-mapped address regions, instead of just text+data+bss.

"tuntap" parameters

The "tuntap" driver configures a point-to-point interface between the vpp engine and the local Linux kernel stack. This allows e.g. users to ssh to the host / VM / container via vpp "revenue" interfaces. It's marginally useful, and is currently disabled by default. To [dynamically] create TAP interfaces - the preferred scheme - see the "tap_connect" binary API

tuntap disable Disable the tun/tap driver if desired
tuntap ethernet Create a tap device (ethernet MAC) instead of a tun device (point-to-point tunnel)
tuntap have-normal-interface Treat the host Linux stack as a routing peer instead of programming vpp interface L3 addresses onto the tun/tap devices. You'll need to manually configure the Linux network stack "vnet" interface, configure vlan subinterfaces if desired, etc.
tuntap name ishmael Call the tun/tap device ishmael

Here's a typical multiple parameter invocation:

 tuntap { ethernet have-normal-interface name vpp1 }

"api-trace" parameters

The ability to trace, dump, and replay control-plane API traces makes all the difference in the world when trying to understand what the control-plane has tried to ask the forwarding-plane to do.

on Enable API trace capture from the beginning of time, and arrange for a post-mortem dump of the API trace if the application terminates abnormally. By default, the (circular) trace buffer will be configured to capture 256K traces
nitems <nnn> Configure the circular trace buffer to contain the last <nnn> entries. By default, the trace buffer captures the last 256K API messages received.

Typically, one simply enables the API message trace scheme:

api-trace { on }

"cpu" parameters

Command-line CPU configuration controls the creation of named thread types, and the cpu affinity thereof. As of this writing, the cpu configuration function supports these parameters:

workers <nn> Create <nn> worker threads.
io <nn> Create nn i/o threads.
main-thread-io Handle i/o devices from thread 0, hand off traffic to worker threads. Requires "workers <nn>".
skip-cores <nn> Leave the low nn bits of the process affinity mask clear.

In some cases, it may be necessary to place thread instances manually:

main-core <n> Assign main thread to specific core.
coremask-workers <hex-mask> Place worker threads according to the bitmap hex-mask.
corelist-workers <list> Same as coremask-workers but accepts a list of cores instead of a bitmap.
coremask-io <hex-mask> Place I/O threads according to the bitmap hex-mask.
corelist-io <list> Same as coremask-io but accepts a list of cores instead of a bitmap.

Note that the "main" thread always occupies the lowest core-id specified in the DPDK [process-level] coremask.

Here's a full-bore manual placement example:

/cisco/bin/vpe  unix interactive tuntap disable cpu { main-thread-io coremask-workers 18 coremask-stats 4 } dpdk {coremask 1e}

# taskset -a -p <vpe-pid>
pid 16251's current affinity mask: 2   # main thread
pid 16288's current affinity mask: ffffff # DPDK interrupt thread (not bound to a core)
pid 16289's current affinity mask: 4 # stats thread
pid 16290's current affinity mask: 8 # worker thread 0
pid 16291's current affinity mask: 10 # worker thread 1

"dpdk" parameters

Command line DPDK configuration controls a number of parameters, including device whitelisting, the number of CPUs available for launching dpdk-eal-controlled threads, the number of I/O buffers, and the process affinity mask.

In addition, the DPDK configuration function attempts to support all of the DPDK EAL configuration parameters.

Popular options include:

coremask <hex> Sets the process-level coremask. See previous section on cpu parameters for more details.
no-hugetlb Don't use huge TLB pages. Potentially useful for running simulator images.
dev <pci-dev> White-list [as in, attempt to drive] a specific PCI device. PCI-dev is a string of the form "DDDD:BB:SS.F" where DDDD = Domain, BB = Bus Number, SS = Slot number, and F = Function. This is the same format used in the linux sysfs tree (i.e. /sys/bus/pci/devices) for PCI device directory names.
vdev <DPDK EAL command to create link bonding interfaces> Specify bonded Ethernet interfaces, operating modes and PCI address of slave links. For example, "vdev eth_bond0,mode=2,slave=0000:0f:00.0,slave=0000:11:00.0,xmit_policy=l34 vdev eth_bond1,mode=2,slave=0000:10:00.0,slave=0000:12:00.0,xmit_policy=l34". Only XOR balanced (mode 2) mode is supported.
num-mbufs <nn> Number of I/O buffers. You need to increase this if you have a lot of physical interfaces with multiple RSS queues (for example, 131072).
kni <nn> Number of KNI interfaces. Refer to the DPDK documentation
uio-driver <name> Defaults to "igb_uio" [the Ubuntu default]. The names "uio_pci_generic" or "vfio-pci" are other possible values.

All of the DPDK EAL options should be available. See .../open-repo/vnet/vnet/devices/dpdk/dpdk.h, look at the set of "foreach_eal_XXX" macros.

"l2tp" parameters

ip6-l2tpv3 configuration controls the method used to locate a specific ip6-l2tpv3 tunnel. The following settings are mutually exclusive

lookup-v6-src Lookup tunnel by ip6 source address.
lookup-v6-dst Lookup tunnel by ip6 destination address.
lookup-session-id Lookup tunnel by l2tpv3 session identifier.

Refer to .../open-repo/vnet/vnet/devices/dpdk/{dpdk.h, init.c} for more details.

"oam" parameters

Oam configuration controls the (ip4-icmp) interval, and number of misses allowed before reporting an oam target down to any registered listener.

interval <floating-point-seconds> Interval between sending oam ip4 icmp messages.
misses-allowed <nn> Number of misses before declaring an oam target down.

"heapsize" parameter

Heapsize configuration controls the size of the main heap. The heap size is configured very early in the boot sequence, before loading plug-ins or doing much of anything else.

<nn>G Specifies the size of the heap in mb or gb. Setting the main heap size to 4gb or more requires recompilation of the entire system with CLIB_VEC64 > 0. See .../clib/clib/vec_bootstrap.h.

"plugin_path" parameter

Plugin_path configuration controls the set of directories searched for vlib plugins. Supply a colon-separated list of (absolute) directory names:

plugin_path dir1:dir2:...:dirN Configure the search path for plugins.

"cj" parameters

The cj thread-safe circular log buffer scheme is occasionally useful when chasing bugs. Calls to it should not be checked in. See .../vlib/vlib/unix/cj.c.

records <nn> Configure the number of cj records in the circular buffer.
on Turns on logging at the earliest possible moment.

"vhost-user" parameters

Vhost-user configuration parameters control the vhost-user driver.

coalesce-frames <nn> Subject to deadline-timer expiration - see next item - attempt to transmit at least nn packet frames.
coalesce-time <seconds> Hold packets no longer than (floating-point) seconds</coode> before transmitting them.
<code>dont-dump-memory vhost-user shared-memory segments can add up to a large amount of memory, so it's handy to avoid adding them to corefiles when using a significant number of such interfaces.

"l2learn" parameter

Configures the number of L2 (MAC) addresses in the L2 FIB at any one time. Defaults to 1M entries.

limit <nn> Limits the size of the L2 FIB to nn concurrent entries.