1 .\" Copyright (c) 2008 Pyun YongHyeon
2 .\" All rights reserved.
4 .\" Redistribution and use in source and binary forms, with or without
5 .\" modification, are permitted provided that the following conditions
7 .\" 1. Redistributions of source code must retain the above copyright
8 .\" notice, this list of conditions and the following disclaimer.
9 .\" 2. Redistributions in binary form must reproduce the above copyright
10 .\" notice, this list of conditions and the following disclaimer in the
11 .\" documentation and/or other materials provided with the distribution.
13 .\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
14 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
15 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
16 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
17 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
18 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
19 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
20 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
21 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
22 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
25 .\" $FreeBSD: src/share/man/man4/jme.4,v 1.1 2008/05/27 01:59:17 yongari Exp $
32 .Nd JMicron Gigabit/Fast Ethernet driver
34 To compile this driver into the kernel,
35 place the following lines in your
36 kernel configuration file:
37 .Bd -ragged -offset indent
42 Alternatively, to load the driver as a
43 module at boot time, place the following line in
45 .Bd -literal -offset indent
51 device driver provides support for JMicron JMC250 PCI Express
52 Gigabit Ethernet controllers and JMicron JMC260 PCI Express Fast
55 All LOMs supported by the
57 driver have TCP/UDP/IP checksum offload for both transmit and receive,
58 TCP segmentation offload (TSO),
59 receive side scaling (RSS),
60 multiple vector MSI-X,
61 hardware VLAN tag stripping/insertion features,
63 an interrupt coalescing/moderation mechanism as well as
64 a 64-bit multicast hash filter.
65 Up to 4 reception queues can be enabled.
68 driver will try enabling as many reception queues as allowed by
69 the number of CPUs in the system.
71 The JMC250 also supports Jumbo Frames (up to 9216 bytes), which can be
72 configured via the interface MTU setting.
73 Selecting an MTU larger than 1500 bytes with the
75 utility configures the adapter to receive and transmit Jumbo Frames.
79 driver supports the following media types:
80 .Bl -tag -width ".Cm 10baseT/UTP"
82 Enable autoselection of the media type and options.
83 The user can manually override
84 the autoselected mode by adding media options to
89 Set 100Mbps (Fast Ethernet) operation.
91 Set 1000baseT operation over twisted pair.
96 driver supports the following media options:
97 .Bl -tag -width ".Cm full-duplex"
99 Force full duplex operation.
101 Force half duplex operation.
104 For more information on configuring this device, see
111 By default, the driver will use MSI-X if it is supported.
112 This behavior can be turned off by setting the following tunable:
114 is the device unit number):
115 .Bd -literal -offset indent
117 hw.jmeX.msix.enable=0
121 the transmission queue MSI-X's target CPU
122 can be specified using the following tunable:
123 .Bd -literal -offset indent
127 The maximum allowed value for this tunable is
128 the power of 2 number of CPUs minus one.
131 the reception queue MSI-Xs' leading target CPU
132 can be specified using the following tunable:
133 .Bd -literal -offset indent
137 The value specificed must be aligned to the number of reception queues enabled
138 and must be less than the power of 2 number of CPUs.
140 If MSI-X is disabled,
141 the driver will use MSI if it is supported.
142 This behavior can be turned off by setting the following tunable:
143 .Bd -literal -offset indent
149 MSI's target CPU can be changed by the following tunable:
150 .Bd -literal -offset indent
154 The maximum allowed value for this tunable is the number of CPUs minus one.
156 Number of reception queues that can be enabled:
157 .Bd -literal -offset indent
159 hw.jmeX.rx_ring_count
162 Maximum allowed value for these tunables is 4
163 and it must be power of 2 aligned.
164 Setting these tunables to 0 allows driver to enable as many reception queues
165 as allowed by the number of CPUs.
167 Number of reception descriptors that can be used:
168 .Bd -literal -offset indent
170 hw.jmeX.rx_desc_count
173 Maximum allowed value for these tunables is 1024.
175 Number of transmission descriptors that can be used:
176 .Bd -literal -offset indent
178 hw.jmeX.tx_desc_count
181 Maximum allowed value for these tunables is 1024.
184 reception queues' leading target CPU can be specified by the following tunable:
185 .Bd -literal -offset indent
189 The value specificed must be aligned to the number of reception queues enabled
190 and must be less than the power of 2 number of CPUs.
193 transmission queue's target CPU can be specified by the following tunable:
194 .Bd -literal -offset indent
198 The maximum allowed value for this tunable is
199 the power of 2 number of CPUs minus one.
201 A number of per-interface variables are implemented in the
206 .Bl -tag -width "rx_ring_count"
208 Maximum amount of time to delay for TX completion interrupt in
210 The accepted range is 1 to 65535, the default is 250 (250us).
212 Maximum number of packets to fire TX completion interrupt.
213 The accepted range is 0 to 255, the default is 128.
214 Packet count based TX interrupt coalescing could be disabled
215 by setting this variable to 0.
217 Maximum amount of time to delay for RX completion interrupt in
219 The accepted range is 1 to 65535, the default is 150 (150us).
221 Maximum number of packets to fire RX completion interrupt.
222 The accepted range is 0 to 255, the default is 64.
223 Packet count based RX interrupt coalescing could be disabled
224 by setting this variable to 0.
226 Number of RX descriptors per-ring (read-only).
228 .Va hw.jme.rx_desc_count
231 Number of TX descriptors (read-only).
233 .Va hw.jme.tx_desc_count
236 Number of RX rings (read-only).
238 .Va hw.jme.rx_ring_count
241 The number of transmission descriptors should be setup
242 before the hardware register is written.
243 Setting this value too high will have negative effect on
244 transmission timeliness.
245 Setting this value too low will hurt overall transmission
246 due to the frequent hardware register writing.
249 reception queues' leading target CPU.
250 It has the same constraints as the tunable
251 .Va hw.jmeX.npoll.rxoff .
252 The set value will take effect the next time
254 is enabled on the device.
257 transmission queue's target CPU.
258 It has the same constraints as the tunable
259 .Va hw.jmeX.npoll.txoff .
260 The set value will take effect the next time
262 is enabled on the device.
277 driver was written by
278 .An Pyun YongHyeon Aq Mt yongari@FreeBSD.org .
281 and was imported into
284 added the receive side scaling and
285 multiple vector MSI-X support to