devnp-e1000.so

Use commas, not spaces, to separate the options.

did=0xXXXX

Detect only devices with this specific PCI Device ID. The default is automatically detected on supported hardware.

duplex=0|1

Half (0) or full (1) duplex mode. The default is automatically detected on supported hardware. If you specify duplex, specify speed as well; if duplex alone is specified, it is ignored and both speed and duplex are auto-negotiated.

You can also use ifconfig -m and ifconfig wmX media to set this.

irq=N

IRQ of the interface. The default is automatically detected on supported hardware.

mac=XXXXXXXXXXXX

The MAC address of the controller. The default is automatically detected on supported hardware.

mtu=N

The maximum transmission unit. The default is 1514.

nomulticast

Disable the driver from sending or receiving multicast packets. By default, multicast is enabled.

pauseignore

Ignore pause frames with respect to full duplex flow control.

pausesuppress

Suppress pause frames with respect to full duplex flow control.

pci=0xXXXX

Detect only devices at this specific PCI index.

priority=N

The priority of the driver's event-handler thread (default 21).

receive=num

The number of receive descriptors; the default is 64.

speed=N

Force the link speed (specified in Mbits/second). If you specify speed, specify duplex as well; if speed alone is specified, the specified speed is correctly set, but duplex defaults to half (0).

You can also use ifconfig -m and ifconfig wmX media to set this.

transmit=N

The number of transmit descriptors; the default is 128.

verbose
verbose=num

Be verbose. Specify num for more verbosity (num can be 1-4; the higher the number, the more detailed the output). The output goes to slogger; invoke sloginfo to view it.

vid=0xXXXX

Detect only devices with this specific PCI Vendor ID.

Description:

The devnp-e1000.so driver manages all current Intel Gigabit devices. This is a native io-pkt driver; its interface names are in the form wmX, where X is an integer.

The devnp-e1000.so and devnp-i82544.so drivers are similar:

devnp-i82544.so has performance optimizations (TSO and interrupt thresholding options) that may make it a better performing candidate for some hardware

devnp-e1000.so likely supports more hardware

If the device enumerators (see enum-devices) don't recognize your device, try explicitly specifying the device ID with the did option when you start the driver.

Some devices support hardware checksums, although some might do so in only one direction; to determine if your device does, type:

ifconfig wmX

and look for the following in the list of supported options:

ip4csum, ip4csum-rx, ip4csum-tx

tcp4csum, tcp4csum-rx, tcp4csum-tx

tcp6csum, tcp6csum-rx, tcp6csum-tx

udp4csum, udp4csum-rx, udp4csum-tx

udp6csum, udp6csum-rx, udp6csum-tx

You can then use ifconfig to enable or disable whichever of these options your device supports.

Native io-pkt and ported NetBSD drivers don't put entries into the /dev/io-net namespace, so a waitfor command for such an entry won't work properly in buildfiles or scripts. Use if_up -p instead; for example, instead of waitfor /dev/io-net/wm0, use if_up -p wm0.

The SQE (Squelch Test Errors) counter — one of the fields reported by nicinfo — isn't applicable to devnp-e1000.so, so this driver uses it in a non-standard way. You can lose packets because:

you ran out of descriptors (the NIC was able to buffer the packet, but there was no CPU RAM available)

or:

the NIC was unable to buffer the packet because it overran its internal Rx FIFO

Other drivers add the two together, but this driver uses the SQE counter for internal Rx FIFO overruns, which generally indicate excessive bus latency, perhaps misconfigured link-level flow control, or even misconfigured Rx FIFO watermarks.

`devnp-e1000.so`

Syntax:

Runs on:

Options:

Description:

Examples:

See also: