| リビジョン | 9aa26dfa4e1584b14c9ee3682c2592708d83c103 (tree) |
|---|---|
| 日時 | 2007-02-06 09:56:59 |
| 作者 | Jesse Brandeburg <jesse.brandeburg@inte...> |
| コミッター | Auke Kok |
e1000: update README
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Documentation/networking/e1000.txt (diff)| @@ -1,7 +1,7 @@ | ||
| 1 | 1 | Linux* Base Driver for the Intel(R) PRO/1000 Family of Adapters |
| 2 | 2 | =============================================================== |
| 3 | 3 | |
| 4 | -November 17, 2004 | |
| 4 | +November 11, 2006 | |
| 5 | 5 | |
| 6 | 6 | |
| 7 | 7 | Contents |
| @@ -9,6 +9,7 @@ Contents | ||
| 9 | 9 | |
| 10 | 10 | - In This Release |
| 11 | 11 | - Identifying Your Adapter |
| 12 | +- Building and Installation | |
| 12 | 13 | - Command Line Parameters |
| 13 | 14 | - Speed and Duplex Configuration |
| 14 | 15 | - Additional Configurations |
| @@ -20,256 +21,398 @@ In This Release | ||
| 20 | 21 | =============== |
| 21 | 22 | |
| 22 | 23 | This file describes the Linux* Base Driver for the Intel(R) PRO/1000 Family |
| 23 | -of Adapters, version 5.x.x. | |
| 24 | +of Adapters. This driver supports kernel versions 2.4.x. This driver includes | |
| 25 | +support for Itanium(R)2-based systems. | |
| 24 | 26 | |
| 25 | -For questions related to hardware requirements, refer to the documentation | |
| 26 | -supplied with your Intel PRO/1000 adapter. All hardware requirements listed | |
| 27 | +For questions related to hardware requirements, refer to the documentation | |
| 28 | +supplied with your Intel PRO/1000 adapter. All hardware requirements listed | |
| 27 | 29 | apply to use with Linux. |
| 28 | 30 | |
| 29 | -Native VLANs are now available with supported kernels. | |
| 31 | +This release includes support for Intel(R) I/O Acceleration Technology, | |
| 32 | +Intel(R) I/OAT. This is supported on systems using the Intel(R) 5000 Series | |
| 33 | +Chipsets Integrated Device - 1A38. You can find additional information | |
| 34 | +on Intel I/OAT at http://www.intel.com/technology/ioacceleration/index.htm. | |
| 35 | + | |
| 36 | +The following features are now available in supported kernels: | |
| 37 | + - Native VLANs | |
| 38 | + - Channel Bonding (teaming) | |
| 39 | + - SNMP | |
| 40 | + | |
| 41 | +Channel Bonding documentation can be found in the Linux kernel source: | |
| 42 | +/Documentation/networking/bonding.txt | |
| 43 | + | |
| 44 | +The driver information previously displayed in the /proc filesystem is not | |
| 45 | +supported in this release. Alternatively, you can use ethtool (version 1.6 | |
| 46 | +or later), lspci, and ifconfig to obtain the same information. | |
| 47 | + | |
| 48 | +Instructions on updating ethtool can be found in the section "Additional | |
| 49 | +Configurations" later in this document. | |
| 50 | + | |
| 51 | +NOTE: The Intel(R) 82562v 10/100 Network Connection only provides 10/100 | |
| 52 | +support. | |
| 53 | + | |
| 30 | 54 | |
| 31 | 55 | Identifying Your Adapter |
| 32 | 56 | ======================== |
| 33 | 57 | |
| 34 | -For more information on how to identify your adapter, go to the Adapter & | |
| 58 | +For more information on how to identify your adapter, go to the Adapter & | |
| 35 | 59 | Driver ID Guide at: |
| 36 | 60 | |
| 37 | 61 | http://support.intel.com/support/network/adapter/pro100/21397.htm |
| 38 | 62 | |
| 39 | -For the latest Intel network drivers for Linux, refer to the following | |
| 40 | -website. In the search field, enter your adapter name or type, or use the | |
| 63 | +For the latest Intel network drivers for Linux, refer to the following | |
| 64 | +website. In the search field, enter your adapter name or type, or use the | |
| 41 | 65 | networking link on the left to search for your adapter: |
| 42 | 66 | |
| 43 | 67 | http://downloadfinder.intel.com/scripts-df/support_intel.asp |
| 44 | 68 | |
| 69 | + | |
| 70 | +Building and Installation | |
| 71 | +========================= | |
| 72 | + | |
| 73 | +This driver is part of the kernel archive you just extracted, be sure to enable | |
| 74 | +the e1000 driver (and probably NAPI in the: | |
| 75 | +Network Device Support | |
| 76 | + --> Ethernet Drivers (1000 Mbit) | |
| 77 | + --> Intel PRO/1000 Gigabit Ethernet Support | |
| 78 | + | |
| 45 | 79 | Command Line Parameters |
| 46 | 80 | ======================= |
| 47 | 81 | |
| 48 | -If the driver is built as a module, the following optional parameters are | |
| 49 | -used by entering them on the command line with the modprobe or insmod command | |
| 82 | +If the driver is built as a module, the following optional parameters | |
| 83 | +are used by entering them on the command line with the modprobe command | |
| 50 | 84 | using this syntax: |
| 51 | 85 | |
| 52 | 86 | modprobe e1000 [<option>=<VAL1>,<VAL2>,...] |
| 53 | 87 | |
| 54 | - insmod e1000 [<option>=<VAL1>,<VAL2>,...] | |
| 55 | - | |
| 56 | 88 | For example, with two PRO/1000 PCI adapters, entering: |
| 57 | 89 | |
| 58 | - insmod e1000 TxDescriptors=80,128 | |
| 90 | + modprobe e1000 TxDescriptors=80,128 | |
| 59 | 91 | |
| 60 | -loads the e1000 driver with 80 TX descriptors for the first adapter and 128 TX | |
| 61 | -descriptors for the second adapter. | |
| 92 | +loads the e1000 driver with 80 TX descriptors for the first adapter and | |
| 93 | +128 TX descriptors for the second adapter. | |
| 62 | 94 | |
| 63 | 95 | The default value for each parameter is generally the recommended setting, |
| 64 | -unless otherwise noted. Also, if the driver is statically built into the | |
| 65 | -kernel, the driver is loaded with the default values for all the parameters. | |
| 66 | -Ethtool can be used to change some of the parameters at runtime. | |
| 96 | +unless otherwise noted. | |
| 67 | 97 | |
| 68 | - NOTES: For more information about the AutoNeg, Duplex, and Speed | |
| 69 | - parameters, see the "Speed and Duplex Configuration" section in | |
| 70 | - this document. | |
| 98 | +NOTES: For more information about the AutoNeg, Duplex, and Speed | |
| 99 | + parameters, see the "Speed and Duplex Configuration" section in | |
| 100 | + this document. | |
| 71 | 101 | |
| 72 | - For more information about the InterruptThrottleRate, RxIntDelay, | |
| 73 | - TxIntDelay, RxAbsIntDelay, and TxAbsIntDelay parameters, see the | |
| 74 | - application note at: | |
| 75 | - http://www.intel.com/design/network/applnots/ap450.htm | |
| 102 | + For more information about the InterruptThrottleRate, | |
| 103 | + RxIntDelay, TxIntDelay, RxAbsIntDelay, and TxAbsIntDelay | |
| 104 | + parameters, see the application note at: | |
| 105 | + http://www.intel.com/design/network/applnots/ap450.htm | |
| 76 | 106 | |
| 77 | - A descriptor describes a data buffer and attributes related to the | |
| 78 | - data buffer. This information is accessed by the hardware. | |
| 107 | + A descriptor describes a data buffer and attributes related to | |
| 108 | + the data buffer. This information is accessed by the hardware. | |
| 79 | 109 | |
| 80 | -AutoNeg (adapters using copper connections only) | |
| 81 | -Valid Range: 0x01-0x0F, 0x20-0x2F | |
| 110 | + | |
| 111 | +AutoNeg | |
| 112 | +------- | |
| 113 | +(Supported only on adapters with copper connections) | |
| 114 | +Valid Range: 0x01-0x0F, 0x20-0x2F | |
| 82 | 115 | Default Value: 0x2F |
| 83 | - This parameter is a bit mask that specifies which speed and duplex | |
| 84 | - settings the board advertises. When this parameter is used, the Speed and | |
| 85 | - Duplex parameters must not be specified. | |
| 86 | - NOTE: Refer to the Speed and Duplex section of this readme for more | |
| 87 | - information on the AutoNeg parameter. | |
| 88 | - | |
| 89 | -Duplex (adapters using copper connections only) | |
| 90 | -Valid Range: 0-2 (0=auto-negotiate, 1=half, 2=full) | |
| 116 | + | |
| 117 | +This parameter is a bit-mask that specifies the speed and duplex settings | |
| 118 | +advertised by the adapter. When this parameter is used, the Speed and | |
| 119 | +Duplex parameters must not be specified. | |
| 120 | + | |
| 121 | +NOTE: Refer to the Speed and Duplex section of this readme for more | |
| 122 | + information on the AutoNeg parameter. | |
| 123 | + | |
| 124 | + | |
| 125 | +Duplex | |
| 126 | +------ | |
| 127 | +(Supported only on adapters with copper connections) | |
| 128 | +Valid Range: 0-2 (0=auto-negotiate, 1=half, 2=full) | |
| 91 | 129 | Default Value: 0 |
| 92 | - Defines the direction in which data is allowed to flow. Can be either one | |
| 93 | - or two-directional. If both Duplex and the link partner are set to auto- | |
| 94 | - negotiate, the board auto-detects the correct duplex. If the link partner | |
| 95 | - is forced (either full or half), Duplex defaults to half-duplex. | |
| 130 | + | |
| 131 | +This defines the direction in which data is allowed to flow. Can be | |
| 132 | +either one or two-directional. If both Duplex and the link partner are | |
| 133 | +set to auto-negotiate, the board auto-detects the correct duplex. If the | |
| 134 | +link partner is forced (either full or half), Duplex defaults to half- | |
| 135 | +duplex. | |
| 136 | + | |
| 96 | 137 | |
| 97 | 138 | FlowControl |
| 98 | -Valid Range: 0-3 (0=none, 1=Rx only, 2=Tx only, 3=Rx&Tx) | |
| 99 | -Default: Read flow control settings from the EEPROM | |
| 100 | - This parameter controls the automatic generation(Tx) and response(Rx) to | |
| 101 | - Ethernet PAUSE frames. | |
| 139 | +----------- | |
| 140 | +Valid Range: 0-3 (0=none, 1=Rx only, 2=Tx only, 3=Rx&Tx) | |
| 141 | +Default Value: Reads flow control settings from the EEPROM | |
| 142 | + | |
| 143 | +This parameter controls the automatic generation(Tx) and response(Rx) | |
| 144 | +to Ethernet PAUSE frames. | |
| 145 | + | |
| 102 | 146 | |
| 103 | 147 | InterruptThrottleRate |
| 104 | -Valid Range: 100-100000 (0=off, 1=dynamic) | |
| 105 | -Default Value: 8000 | |
| 106 | - This value represents the maximum number of interrupts per second the | |
| 107 | - controller generates. InterruptThrottleRate is another setting used in | |
| 108 | - interrupt moderation. Dynamic mode uses a heuristic algorithm to adjust | |
| 109 | - InterruptThrottleRate based on the current traffic load. | |
| 110 | -Un-supported Adapters: InterruptThrottleRate is NOT supported by 82542, 82543 | |
| 111 | - or 82544-based adapters. | |
| 112 | - | |
| 113 | - NOTE: InterruptThrottleRate takes precedence over the TxAbsIntDelay and | |
| 114 | - RxAbsIntDelay parameters. In other words, minimizing the receive | |
| 115 | - and/or transmit absolute delays does not force the controller to | |
| 116 | - generate more interrupts than what the Interrupt Throttle Rate | |
| 117 | - allows. | |
| 118 | - CAUTION: If you are using the Intel PRO/1000 CT Network Connection | |
| 119 | - (controller 82547), setting InterruptThrottleRate to a value | |
| 120 | - greater than 75,000, may hang (stop transmitting) adapters under | |
| 121 | - certain network conditions. If this occurs a NETDEV WATCHDOG | |
| 122 | - message is logged in the system event log. In addition, the | |
| 123 | - controller is automatically reset, restoring the network | |
| 124 | - connection. To eliminate the potential for the hang, ensure | |
| 125 | - that InterruptThrottleRate is set no greater than 75,000 and is | |
| 126 | - not set to 0. | |
| 127 | - NOTE: When e1000 is loaded with default settings and multiple adapters are | |
| 128 | - in use simultaneously, the CPU utilization may increase non-linearly. | |
| 129 | - In order to limit the CPU utilization without impacting the overall | |
| 130 | - throughput, we recommend that you load the driver as follows: | |
| 131 | - | |
| 132 | - insmod e1000.o InterruptThrottleRate=3000,3000,3000 | |
| 133 | - | |
| 134 | - This sets the InterruptThrottleRate to 3000 interrupts/sec for the | |
| 135 | - first, second, and third instances of the driver. The range of 2000 to | |
| 136 | - 3000 interrupts per second works on a majority of systems and is a | |
| 137 | - good starting point, but the optimal value will be platform-specific. | |
| 138 | - If CPU utilization is not a concern, use RX_POLLING (NAPI) and default | |
| 139 | - driver settings. | |
| 148 | +--------------------- | |
| 149 | +(not supported on Intel(R) 82542, 82543 or 82544-based adapters) | |
| 150 | +Valid Range: 0,1,3,100-100000 (0=off, 1=dynamic, 3=dynamic conservative) | |
| 151 | +Default Value: 3 | |
| 152 | + | |
| 153 | +The driver can limit the amount of interrupts per second that the adapter | |
| 154 | +will generate for incoming packets. It does this by writing a value to the | |
| 155 | +adapter that is based on the maximum amount of interrupts that the adapter | |
| 156 | +will generate per second. | |
| 157 | + | |
| 158 | +Setting InterruptThrottleRate to a value greater or equal to 100 | |
| 159 | +will program the adapter to send out a maximum of that many interrupts | |
| 160 | +per second, even if more packets have come in. This reduces interrupt | |
| 161 | +load on the system and can lower CPU utilization under heavy load, | |
| 162 | +but will increase latency as packets are not processed as quickly. | |
| 163 | + | |
| 164 | +The default behaviour of the driver previously assumed a static | |
| 165 | +InterruptThrottleRate value of 8000, providing a good fallback value for | |
| 166 | +all traffic types,but lacking in small packet performance and latency. | |
| 167 | +The hardware can handle many more small packets per second however, and | |
| 168 | +for this reason an adaptive interrupt moderation algorithm was implemented. | |
| 169 | + | |
| 170 | +Since 7.3.x, the driver has two adaptive modes (setting 1 or 3) in which | |
| 171 | +it dynamically adjusts the InterruptThrottleRate value based on the traffic | |
| 172 | +that it receives. After determining the type of incoming traffic in the last | |
| 173 | +timeframe, it will adjust the InterruptThrottleRate to an appropriate value | |
| 174 | +for that traffic. | |
| 175 | + | |
| 176 | +The algorithm classifies the incoming traffic every interval into | |
| 177 | +classes. Once the class is determined, the InterruptThrottleRate value is | |
| 178 | +adjusted to suit that traffic type the best. There are three classes defined: | |
| 179 | +"Bulk traffic", for large amounts of packets of normal size; "Low latency", | |
| 180 | +for small amounts of traffic and/or a significant percentage of small | |
| 181 | +packets; and "Lowest latency", for almost completely small packets or | |
| 182 | +minimal traffic. | |
| 183 | + | |
| 184 | +In dynamic conservative mode, the InterruptThrottleRate value is set to 4000 | |
| 185 | +for traffic that falls in class "Bulk traffic". If traffic falls in the "Low | |
| 186 | +latency" or "Lowest latency" class, the InterruptThrottleRate is increased | |
| 187 | +stepwise to 20000. This default mode is suitable for most applications. | |
| 188 | + | |
| 189 | +For situations where low latency is vital such as cluster or | |
| 190 | +grid computing, the algorithm can reduce latency even more when | |
| 191 | +InterruptThrottleRate is set to mode 1. In this mode, which operates | |
| 192 | +the same as mode 3, the InterruptThrottleRate will be increased stepwise to | |
| 193 | +70000 for traffic in class "Lowest latency". | |
| 194 | + | |
| 195 | +Setting InterruptThrottleRate to 0 turns off any interrupt moderation | |
| 196 | +and may improve small packet latency, but is generally not suitable | |
| 197 | +for bulk throughput traffic. | |
| 198 | + | |
| 199 | +NOTE: InterruptThrottleRate takes precedence over the TxAbsIntDelay and | |
| 200 | + RxAbsIntDelay parameters. In other words, minimizing the receive | |
| 201 | + and/or transmit absolute delays does not force the controller to | |
| 202 | + generate more interrupts than what the Interrupt Throttle Rate | |
| 203 | + allows. | |
| 204 | + | |
| 205 | +CAUTION: If you are using the Intel(R) PRO/1000 CT Network Connection | |
| 206 | + (controller 82547), setting InterruptThrottleRate to a value | |
| 207 | + greater than 75,000, may hang (stop transmitting) adapters | |
| 208 | + under certain network conditions. If this occurs a NETDEV | |
| 209 | + WATCHDOG message is logged in the system event log. In | |
| 210 | + addition, the controller is automatically reset, restoring | |
| 211 | + the network connection. To eliminate the potential for the | |
| 212 | + hang, ensure that InterruptThrottleRate is set no greater | |
| 213 | + than 75,000 and is not set to 0. | |
| 214 | + | |
| 215 | +NOTE: When e1000 is loaded with default settings and multiple adapters | |
| 216 | + are in use simultaneously, the CPU utilization may increase non- | |
| 217 | + linearly. In order to limit the CPU utilization without impacting | |
| 218 | + the overall throughput, we recommend that you load the driver as | |
| 219 | + follows: | |
| 220 | + | |
| 221 | + modprobe e1000 InterruptThrottleRate=3000,3000,3000 | |
| 222 | + | |
| 223 | + This sets the InterruptThrottleRate to 3000 interrupts/sec for | |
| 224 | + the first, second, and third instances of the driver. The range | |
| 225 | + of 2000 to 3000 interrupts per second works on a majority of | |
| 226 | + systems and is a good starting point, but the optimal value will | |
| 227 | + be platform-specific. If CPU utilization is not a concern, use | |
| 228 | + RX_POLLING (NAPI) and default driver settings. | |
| 229 | + | |
| 230 | + | |
| 140 | 231 | |
| 141 | 232 | RxDescriptors |
| 142 | -Valid Range: 80-256 for 82542 and 82543-based adapters | |
| 143 | - 80-4096 for all other supported adapters | |
| 233 | +------------- | |
| 234 | +Valid Range: 80-256 for 82542 and 82543-based adapters | |
| 235 | + 80-4096 for all other supported adapters | |
| 144 | 236 | Default Value: 256 |
| 145 | - This value is the number of receive descriptors allocated by the driver. | |
| 146 | - Increasing this value allows the driver to buffer more incoming packets. | |
| 147 | - Each descriptor is 16 bytes. A receive buffer is allocated for each | |
| 148 | - descriptor and can either be 2048 or 4096 bytes long, depending on the MTU | |
| 149 | 237 | |
| 150 | - setting. An incoming packet can span one or more receive descriptors. | |
| 151 | - The maximum MTU size is 16110. | |
| 238 | +This value specifies the number of receive buffer descriptors allocated | |
| 239 | +by the driver. Increasing this value allows the driver to buffer more | |
| 240 | +incoming packets, at the expense of increased system memory utilization. | |
| 241 | + | |
| 242 | +Each descriptor is 16 bytes. A receive buffer is also allocated for each | |
| 243 | +descriptor and can be either 2048, 4096, 8192, or 16384 bytes, depending | |
| 244 | +on the MTU setting. The maximum MTU size is 16110. | |
| 245 | + | |
| 246 | +NOTE: MTU designates the frame size. It only needs to be set for Jumbo | |
| 247 | + Frames. Depending on the available system resources, the request | |
| 248 | + for a higher number of receive descriptors may be denied. In this | |
| 249 | + case, use a lower number. | |
| 152 | 250 | |
| 153 | - NOTE: MTU designates the frame size. It only needs to be set for Jumbo | |
| 154 | - Frames. | |
| 155 | - NOTE: Depending on the available system resources, the request for a | |
| 156 | - higher number of receive descriptors may be denied. In this case, | |
| 157 | - use a lower number. | |
| 158 | 251 | |
| 159 | 252 | RxIntDelay |
| 160 | -Valid Range: 0-65535 (0=off) | |
| 253 | +---------- | |
| 254 | +Valid Range: 0-65535 (0=off) | |
| 161 | 255 | Default Value: 0 |
| 162 | - This value delays the generation of receive interrupts in units of 1.024 | |
| 163 | - microseconds. Receive interrupt reduction can improve CPU efficiency if | |
| 164 | - properly tuned for specific network traffic. Increasing this value adds | |
| 165 | - extra latency to frame reception and can end up decreasing the throughput | |
| 166 | - of TCP traffic. If the system is reporting dropped receives, this value | |
| 167 | - may be set too high, causing the driver to run out of available receive | |
| 168 | - descriptors. | |
| 169 | - | |
| 170 | - CAUTION: When setting RxIntDelay to a value other than 0, adapters may | |
| 171 | - hang (stop transmitting) under certain network conditions. If | |
| 172 | - this occurs a NETDEV WATCHDOG message is logged in the system | |
| 173 | - event log. In addition, the controller is automatically reset, | |
| 174 | - restoring the network connection. To eliminate the potential for | |
| 175 | - the hang ensure that RxIntDelay is set to 0. | |
| 176 | - | |
| 177 | -RxAbsIntDelay (82540, 82545 and later adapters only) | |
| 178 | -Valid Range: 0-65535 (0=off) | |
| 256 | + | |
| 257 | +This value delays the generation of receive interrupts in units of 1.024 | |
| 258 | +microseconds. Receive interrupt reduction can improve CPU efficiency if | |
| 259 | +properly tuned for specific network traffic. Increasing this value adds | |
| 260 | +extra latency to frame reception and can end up decreasing the throughput | |
| 261 | +of TCP traffic. If the system is reporting dropped receives, this value | |
| 262 | +may be set too high, causing the driver to run out of available receive | |
| 263 | +descriptors. | |
| 264 | + | |
| 265 | +CAUTION: When setting RxIntDelay to a value other than 0, adapters may | |
| 266 | + hang (stop transmitting) under certain network conditions. If | |
| 267 | + this occurs a NETDEV WATCHDOG message is logged in the system | |
| 268 | + event log. In addition, the controller is automatically reset, | |
| 269 | + restoring the network connection. To eliminate the potential | |
| 270 | + for the hang ensure that RxIntDelay is set to 0. | |
| 271 | + | |
| 272 | + | |
| 273 | +RxAbsIntDelay | |
| 274 | +------------- | |
| 275 | +(This parameter is supported only on 82540, 82545 and later adapters.) | |
| 276 | +Valid Range: 0-65535 (0=off) | |
| 179 | 277 | Default Value: 128 |
| 180 | - This value, in units of 1.024 microseconds, limits the delay in which a | |
| 181 | - receive interrupt is generated. Useful only if RxIntDelay is non-zero, | |
| 182 | - this value ensures that an interrupt is generated after the initial | |
| 183 | - packet is received within the set amount of time. Proper tuning, | |
| 184 | - along with RxIntDelay, may improve traffic throughput in specific network | |
| 185 | - conditions. | |
| 186 | - | |
| 187 | -Speed (adapters using copper connections only) | |
| 278 | + | |
| 279 | +This value, in units of 1.024 microseconds, limits the delay in which a | |
| 280 | +receive interrupt is generated. Useful only if RxIntDelay is non-zero, | |
| 281 | +this value ensures that an interrupt is generated after the initial | |
| 282 | +packet is received within the set amount of time. Proper tuning, | |
| 283 | +along with RxIntDelay, may improve traffic throughput in specific network | |
| 284 | +conditions. | |
| 285 | + | |
| 286 | + | |
| 287 | +Speed | |
| 288 | +----- | |
| 289 | +(This parameter is supported only on adapters with copper connections.) | |
| 188 | 290 | Valid Settings: 0, 10, 100, 1000 |
| 189 | -Default Value: 0 (auto-negotiate at all supported speeds) | |
| 190 | - Speed forces the line speed to the specified value in megabits per second | |
| 191 | - (Mbps). If this parameter is not specified or is set to 0 and the link | |
| 192 | - partner is set to auto-negotiate, the board will auto-detect the correct | |
| 193 | - speed. Duplex should also be set when Speed is set to either 10 or 100. | |
| 291 | +Default Value: 0 (auto-negotiate at all supported speeds) | |
| 292 | + | |
| 293 | +Speed forces the line speed to the specified value in megabits per second | |
| 294 | +(Mbps). If this parameter is not specified or is set to 0 and the link | |
| 295 | +partner is set to auto-negotiate, the board will auto-detect the correct | |
| 296 | +speed. Duplex should also be set when Speed is set to either 10 or 100. | |
| 297 | + | |
| 194 | 298 | |
| 195 | 299 | TxDescriptors |
| 196 | -Valid Range: 80-256 for 82542 and 82543-based adapters | |
| 197 | - 80-4096 for all other supported adapters | |
| 300 | +------------- | |
| 301 | +Valid Range: 80-256 for 82542 and 82543-based adapters | |
| 302 | + 80-4096 for all other supported adapters | |
| 198 | 303 | Default Value: 256 |
| 199 | - This value is the number of transmit descriptors allocated by the driver. | |
| 200 | - Increasing this value allows the driver to queue more transmits. Each | |
| 201 | - descriptor is 16 bytes. | |
| 202 | 304 | |
| 203 | - NOTE: Depending on the available system resources, the request for a | |
| 204 | - higher number of transmit descriptors may be denied. In this case, | |
| 205 | - use a lower number. | |
| 305 | +This value is the number of transmit descriptors allocated by the driver. | |
| 306 | +Increasing this value allows the driver to queue more transmits. Each | |
| 307 | +descriptor is 16 bytes. | |
| 308 | + | |
| 309 | +NOTE: Depending on the available system resources, the request for a | |
| 310 | + higher number of transmit descriptors may be denied. In this case, | |
| 311 | + use a lower number. | |
| 312 | + | |
| 206 | 313 | |
| 207 | 314 | TxIntDelay |
| 208 | -Valid Range: 0-65535 (0=off) | |
| 315 | +---------- | |
| 316 | +Valid Range: 0-65535 (0=off) | |
| 209 | 317 | Default Value: 64 |
| 210 | - This value delays the generation of transmit interrupts in units of | |
| 211 | - 1.024 microseconds. Transmit interrupt reduction can improve CPU | |
| 212 | - efficiency if properly tuned for specific network traffic. If the | |
| 213 | - system is reporting dropped transmits, this value may be set too high | |
| 214 | - causing the driver to run out of available transmit descriptors. | |
| 215 | - | |
| 216 | -TxAbsIntDelay (82540, 82545 and later adapters only) | |
| 217 | -Valid Range: 0-65535 (0=off) | |
| 318 | + | |
| 319 | +This value delays the generation of transmit interrupts in units of | |
| 320 | +1.024 microseconds. Transmit interrupt reduction can improve CPU | |
| 321 | +efficiency if properly tuned for specific network traffic. If the | |
| 322 | +system is reporting dropped transmits, this value may be set too high | |
| 323 | +causing the driver to run out of available transmit descriptors. | |
| 324 | + | |
| 325 | + | |
| 326 | +TxAbsIntDelay | |
| 327 | +------------- | |
| 328 | +(This parameter is supported only on 82540, 82545 and later adapters.) | |
| 329 | +Valid Range: 0-65535 (0=off) | |
| 218 | 330 | Default Value: 64 |
| 219 | - This value, in units of 1.024 microseconds, limits the delay in which a | |
| 220 | - transmit interrupt is generated. Useful only if TxIntDelay is non-zero, | |
| 221 | - this value ensures that an interrupt is generated after the initial | |
| 222 | - packet is sent on the wire within the set amount of time. Proper tuning, | |
| 223 | - along with TxIntDelay, may improve traffic throughput in specific | |
| 224 | - network conditions. | |
| 225 | - | |
| 226 | -XsumRX (not available on the 82542-based adapter) | |
| 227 | -Valid Range: 0-1 | |
| 331 | + | |
| 332 | +This value, in units of 1.024 microseconds, limits the delay in which a | |
| 333 | +transmit interrupt is generated. Useful only if TxIntDelay is non-zero, | |
| 334 | +this value ensures that an interrupt is generated after the initial | |
| 335 | +packet is sent on the wire within the set amount of time. Proper tuning, | |
| 336 | +along with TxIntDelay, may improve traffic throughput in specific | |
| 337 | +network conditions. | |
| 338 | + | |
| 339 | +XsumRX | |
| 340 | +------ | |
| 341 | +(This parameter is NOT supported on the 82542-based adapter.) | |
| 342 | +Valid Range: 0-1 | |
| 228 | 343 | Default Value: 1 |
| 229 | - A value of '1' indicates that the driver should enable IP checksum | |
| 230 | - offload for received packets (both UDP and TCP) to the adapter hardware. | |
| 344 | + | |
| 345 | +A value of '1' indicates that the driver should enable IP checksum | |
| 346 | +offload for received packets (both UDP and TCP) to the adapter hardware. | |
| 347 | + | |
| 348 | +Copybreak | |
| 349 | +--------- | |
| 350 | +Valid Range: 0-xxxxxxx | |
| 351 | +Default Value: 256 | |
| 352 | +Usage: insmod e1000.ko copybreak=128 | |
| 353 | + | |
| 354 | +Driver copies all packets below or equaling this size to a fresh rx | |
| 355 | +buffer before handing it up the stack. | |
| 356 | + | |
| 357 | +This parameter is different than other parameters, in that it is a | |
| 358 | +single (not 1,1,1 etc.) parameter applied to all driver instances and | |
| 359 | +it is also available during runtime at | |
| 360 | +/sys/module/e1000/parameters/copybreak | |
| 361 | + | |
| 231 | 362 | |
| 232 | 363 | Speed and Duplex Configuration |
| 233 | 364 | ============================== |
| 234 | 365 | |
| 235 | -Three keywords are used to control the speed and duplex configuration. These | |
| 236 | -keywords are Speed, Duplex, and AutoNeg. | |
| 366 | +Three keywords are used to control the speed and duplex configuration. | |
| 367 | +These keywords are Speed, Duplex, and AutoNeg. | |
| 237 | 368 | |
| 238 | -If the board uses a fiber interface, these keywords are ignored, and the | |
| 369 | +If the board uses a fiber interface, these keywords are ignored, and the | |
| 239 | 370 | fiber interface board only links at 1000 Mbps full-duplex. |
| 240 | 371 | |
| 241 | 372 | For copper-based boards, the keywords interact as follows: |
| 242 | 373 | |
| 243 | - The default operation is auto-negotiate. The board advertises all supported | |
| 244 | - speed and duplex combinations, and it links at the highest common speed and | |
| 245 | - duplex mode IF the link partner is set to auto-negotiate. | |
| 374 | + The default operation is auto-negotiate. The board advertises all | |
| 375 | + supported speed and duplex combinations, and it links at the highest | |
| 376 | + common speed and duplex mode IF the link partner is set to auto-negotiate. | |
| 377 | + | |
| 378 | + If Speed = 1000, limited auto-negotiation is enabled and only 1000 Mbps | |
| 379 | + is advertised (The 1000BaseT spec requires auto-negotiation.) | |
| 246 | 380 | |
| 247 | - If Speed = 1000, limited auto-negotiation is enabled and only 1000 Mbps is | |
| 248 | - advertised (The 1000BaseT spec requires auto-negotiation.) | |
| 381 | + If Speed = 10 or 100, then both Speed and Duplex should be set. Auto- | |
| 382 | + negotiation is disabled, and the AutoNeg parameter is ignored. Partner | |
| 383 | + SHOULD also be forced. | |
| 249 | 384 | |
| 250 | - If Speed = 10 or 100, then both Speed and Duplex should be set. Auto- | |
| 251 | - negotiation is disabled, and the AutoNeg parameter is ignored. Partner SHOULD | |
| 252 | - also be forced. | |
| 385 | +The AutoNeg parameter is used when more control is required over the | |
| 386 | +auto-negotiation process. It should be used when you wish to control which | |
| 387 | +speed and duplex combinations are advertised during the auto-negotiation | |
| 388 | +process. | |
| 253 | 389 | |
| 254 | -The AutoNeg parameter is used when more control is required over the auto- | |
| 255 | -negotiation process. When this parameter is used, Speed and Duplex parameters | |
| 256 | -must not be specified. The following table describes supported values for the | |
| 257 | -AutoNeg parameter: | |
| 390 | +The parameter may be specified as either a decimal or hexadecimal value as | |
| 391 | +determined by the bitmap below. | |
| 258 | 392 | |
| 393 | +Bit position 7 6 5 4 3 2 1 0 | |
| 394 | +Decimal Value 128 64 32 16 8 4 2 1 | |
| 395 | +Hex value 80 40 20 10 8 4 2 1 | |
| 396 | +Speed (Mbps) N/A N/A 1000 N/A 100 100 10 10 | |
| 397 | +Duplex Full Full Half Full Half | |
| 259 | 398 | |
| 260 | -Speed (Mbps) 1000 100 100 10 10 | |
| 261 | -Duplex Full Full Half Full Half | |
| 262 | -Value (in base 16) 0x20 0x08 0x04 0x02 0x01 | |
| 399 | +Some examples of using AutoNeg: | |
| 263 | 400 | |
| 401 | + modprobe e1000 AutoNeg=0x01 (Restricts autonegotiation to 10 Half) | |
| 402 | + modprobe e1000 AutoNeg=1 (Same as above) | |
| 403 | + modprobe e1000 AutoNeg=0x02 (Restricts autonegotiation to 10 Full) | |
| 404 | + modprobe e1000 AutoNeg=0x03 (Restricts autonegotiation to 10 Half or 10 Full) | |
| 405 | + modprobe e1000 AutoNeg=0x04 (Restricts autonegotiation to 100 Half) | |
| 406 | + modprobe e1000 AutoNeg=0x05 (Restricts autonegotiation to 10 Half or 100 | |
| 407 | + Half) | |
| 408 | + modprobe e1000 AutoNeg=0x020 (Restricts autonegotiation to 1000 Full) | |
| 409 | + modprobe e1000 AutoNeg=32 (Same as above) | |
| 264 | 410 | |
| 265 | -Example: insmod e1000 AutoNeg=0x03, loads e1000 and specifies (10 full duplex, | |
| 266 | -10 half duplex) for negotiation with the peer. | |
| 411 | +Note that when this parameter is used, Speed and Duplex must not be specified. | |
| 267 | 412 | |
| 268 | -Note that setting AutoNeg does not guarantee that the board will link at the | |
| 269 | -highest specified speed or duplex mode, but the board will link at the | |
| 270 | -highest possible speed/duplex of the link partner IF the link partner is also | |
| 271 | -set to auto-negotiate. If the link partner is forced speed/duplex, the | |
| 272 | -adapter MUST be forced to the same speed/duplex. | |
| 413 | +If the link partner is forced to a specific speed and duplex, then this | |
| 414 | +parameter should not be used. Instead, use the Speed and Duplex parameters | |
| 415 | +previously mentioned to force the adapter to the same speed and duplex. | |
| 273 | 416 | |
| 274 | 417 | |
| 275 | 418 | Additional Configurations |
| @@ -277,20 +420,19 @@ Additional Configurations | ||
| 277 | 420 | |
| 278 | 421 | Configuring the Driver on Different Distributions |
| 279 | 422 | ------------------------------------------------- |
| 280 | - | |
| 281 | - Configuring a network driver to load properly when the system is started is | |
| 282 | - distribution dependent. Typically, the configuration process involves adding | |
| 283 | - an alias line to /etc/modules.conf as well as editing other system startup | |
| 284 | - scripts and/or configuration files. Many popular Linux distributions ship | |
| 285 | - with tools to make these changes for you. To learn the proper way to | |
| 286 | - configure a network device for your system, refer to your distribution | |
| 287 | - documentation. If during this process you are asked for the driver or module | |
| 288 | - name, the name for the Linux Base Driver for the Intel PRO/1000 Family of | |
| 289 | - Adapters is e1000. | |
| 290 | - | |
| 291 | - As an example, if you install the e1000 driver for two PRO/1000 adapters | |
| 292 | - (eth0 and eth1) and set the speed and duplex to 10full and 100half, add the | |
| 293 | - following to modules.conf: | |
| 423 | + Configuring a network driver to load properly when the system is started | |
| 424 | + is distribution dependent. Typically, the configuration process involves | |
| 425 | + adding an alias line to /etc/modules.conf or /etc/modprobe.conf as well | |
| 426 | + as editing other system startup scripts and/or configuration files. Many | |
| 427 | + popular Linux distributions ship with tools to make these changes for you. | |
| 428 | + To learn the proper way to configure a network device for your system, | |
| 429 | + refer to your distribution documentation. If during this process you are | |
| 430 | + asked for the driver or module name, the name for the Linux Base Driver | |
| 431 | + for the Intel(R) PRO/1000 Family of Adapters is e1000. | |
| 432 | + | |
| 433 | + As an example, if you install the e1000 driver for two PRO/1000 adapters | |
| 434 | + (eth0 and eth1) and set the speed and duplex to 10full and 100half, add | |
| 435 | + the following to modules.conf or or modprobe.conf: | |
| 294 | 436 | |
| 295 | 437 | alias eth0 e1000 |
| 296 | 438 | alias eth1 e1000 |
| @@ -298,10 +440,9 @@ Additional Configurations | ||
| 298 | 440 | |
| 299 | 441 | Viewing Link Messages |
| 300 | 442 | --------------------- |
| 301 | - | |
| 302 | - Link messages will not be displayed to the console if the distribution is | |
| 303 | - restricting system messages. In order to see network driver link messages on | |
| 304 | - your console, set dmesg to eight by entering the following: | |
| 443 | + Link messages will not be displayed to the console if the distribution is | |
| 444 | + restricting system messages. In order to see network driver link messages | |
| 445 | + on your console, set dmesg to eight by entering the following: | |
| 305 | 446 | |
| 306 | 447 | dmesg -n 8 |
| 307 | 448 |
| @@ -309,58 +450,108 @@ Additional Configurations | ||
| 309 | 450 | |
| 310 | 451 | Jumbo Frames |
| 311 | 452 | ------------ |
| 453 | + Jumbo Frames support is enabled by changing the MTU to a value larger than | |
| 454 | + the default of 1500. Use the ifconfig command to increase the MTU size. | |
| 455 | + For example: | |
| 456 | + | |
| 457 | + ifconfig eth<x> mtu 9000 up | |
| 458 | + | |
| 459 | + This setting is not saved across reboots. It can be made permanent if | |
| 460 | + you add: | |
| 461 | + | |
| 462 | + MTU=9000 | |
| 463 | + | |
| 464 | + to the file /etc/sysconfig/network-scripts/ifcfg-eth<x>. This example | |
| 465 | + applies to the Red Hat distributions; other distributions may store this | |
| 466 | + setting in a different location. | |
| 467 | + | |
| 468 | + Notes: | |
| 469 | + | |
| 470 | + - To enable Jumbo Frames, increase the MTU size on the interface beyond | |
| 471 | + 1500. | |
| 472 | + | |
| 473 | + - The maximum MTU setting for Jumbo Frames is 16110. This value coincides | |
| 474 | + with the maximum Jumbo Frames size of 16128. | |
| 475 | + | |
| 476 | + - Using Jumbo Frames at 10 or 100 Mbps may result in poor performance or | |
| 477 | + loss of link. | |
| 478 | + | |
| 479 | + - Some Intel gigabit adapters that support Jumbo Frames have a frame size | |
| 480 | + limit of 9238 bytes, with a corresponding MTU size limit of 9216 bytes. | |
| 481 | + The adapters with this limitation are based on the Intel(R) 82571EB, | |
| 482 | + 82572EI, 82573L and 80003ES2LAN controller. These correspond to the | |
| 483 | + following product names: | |
| 484 | + Intel(R) PRO/1000 PT Server Adapter | |
| 485 | + Intel(R) PRO/1000 PT Desktop Adapter | |
| 486 | + Intel(R) PRO/1000 PT Network Connection | |
| 487 | + Intel(R) PRO/1000 PT Dual Port Server Adapter | |
| 488 | + Intel(R) PRO/1000 PT Dual Port Network Connection | |
| 489 | + Intel(R) PRO/1000 PF Server Adapter | |
| 490 | + Intel(R) PRO/1000 PF Network Connection | |
| 491 | + Intel(R) PRO/1000 PF Dual Port Server Adapter | |
| 492 | + Intel(R) PRO/1000 PB Server Connection | |
| 493 | + Intel(R) PRO/1000 PL Network Connection | |
| 494 | + Intel(R) PRO/1000 EB Network Connection with I/O Acceleration | |
| 495 | + Intel(R) PRO/1000 EB Backplane Connection with I/O Acceleration | |
| 496 | + Intel(R) PRO/1000 PT Quad Port Server Adapter | |
| 497 | + | |
| 498 | + - Adapters based on the Intel(R) 82542 and 82573V/E controller do not | |
| 499 | + support Jumbo Frames. These correspond to the following product names: | |
| 500 | + Intel(R) PRO/1000 Gigabit Server Adapter | |
| 501 | + Intel(R) PRO/1000 PM Network Connection | |
| 502 | + | |
| 503 | + - The following adapters do not support Jumbo Frames: | |
| 504 | + Intel(R) 82562V 10/100 Network Connection | |
| 505 | + Intel(R) 82566DM Gigabit Network Connection | |
| 506 | + Intel(R) 82566DC Gigabit Network Connection | |
| 507 | + Intel(R) 82566MM Gigabit Network Connection | |
| 508 | + Intel(R) 82566MC Gigabit Network Connection | |
| 509 | + Intel(R) 82562GT 10/100 Network Connection | |
| 510 | + Intel(R) 82562G 10/100 Network Connection | |
| 312 | 511 | |
| 313 | - The driver supports Jumbo Frames for all adapters except 82542-based | |
| 314 | - adapters. Jumbo Frames support is enabled by changing the MTU to a value | |
| 315 | - larger than the default of 1500. Use the ifconfig command to increase the | |
| 316 | - MTU size. For example: | |
| 317 | - | |
| 318 | - ifconfig ethx mtu 9000 up | |
| 319 | - | |
| 320 | - The maximum MTU setting for Jumbo Frames is 16110. This value coincides | |
| 321 | - with the maximum Jumbo Frames size of 16128. | |
| 322 | - | |
| 323 | - NOTE: Jumbo Frames are supported at 1000 Mbps only. Using Jumbo Frames at | |
| 324 | - 10 or 100 Mbps may result in poor performance or loss of link. | |
| 325 | - | |
| 326 | - | |
| 327 | - NOTE: MTU designates the frame size. To enable Jumbo Frames, increase the | |
| 328 | - MTU size on the interface beyond 1500. | |
| 329 | 512 | |
| 330 | 513 | Ethtool |
| 331 | 514 | ------- |
| 332 | - | |
| 333 | 515 | The driver utilizes the ethtool interface for driver configuration and |
| 334 | 516 | diagnostics, as well as displaying statistical information. Ethtool |
| 335 | 517 | version 1.6 or later is required for this functionality. |
| 336 | 518 | |
| 337 | 519 | The latest release of ethtool can be found from |
| 338 | - http://sf.net/projects/gkernel. | |
| 520 | + http://sourceforge.net/projects/gkernel. | |
| 339 | 521 | |
| 340 | - NOTE: Ethtool 1.6 only supports a limited set of ethtool options. Support | |
| 341 | - for a more complete ethtool feature set can be enabled by upgrading | |
| 342 | - ethtool to ethtool-1.8.1. | |
| 522 | + NOTE: Ethtool 1.6 only supports a limited set of ethtool options. Support | |
| 523 | + for a more complete ethtool feature set can be enabled by upgrading | |
| 524 | + to the latest version. | |
| 343 | 525 | |
| 344 | 526 | Enabling Wake on LAN* (WoL) |
| 345 | 527 | --------------------------- |
| 346 | - | |
| 347 | - WoL is configured through the Ethtool* utility. Ethtool is included with | |
| 348 | - all versions of Red Hat after Red Hat 7.2. For other Linux distributions, | |
| 349 | - download and install Ethtool from the following website: | |
| 528 | + WoL is configured through the Ethtool* utility. Ethtool is included with | |
| 529 | + all versions of Red Hat after Red Hat 7.2. For other Linux distributions, | |
| 530 | + download and install Ethtool from the following website: | |
| 350 | 531 | http://sourceforge.net/projects/gkernel. |
| 351 | 532 | |
| 352 | - For instructions on enabling WoL with Ethtool, refer to the website listed | |
| 533 | + For instructions on enabling WoL with Ethtool, refer to the website listed | |
| 353 | 534 | above. |
| 354 | 535 | |
| 355 | - WoL will be enabled on the system during the next shut down or reboot. | |
| 356 | - For this driver version, in order to enable WoL, the e1000 driver must be | |
| 536 | + WoL will be enabled on the system during the next shut down or reboot. | |
| 537 | + For this driver version, in order to enable WoL, the e1000 driver must be | |
| 357 | 538 | loaded when shutting down or rebooting the system. |
| 358 | 539 | |
| 540 | + Wake On LAN is only supported on port A for the following devices: | |
| 541 | + Intel(R) PRO/1000 PT Dual Port Network Connection | |
| 542 | + Intel(R) PRO/1000 PT Dual Port Server Connection | |
| 543 | + Intel(R) PRO/1000 PT Dual Port Server Adapter | |
| 544 | + Intel(R) PRO/1000 PF Dual Port Server Adapter | |
| 545 | + Intel(R) PRO/1000 PT Quad Port Server Adapter | |
| 546 | + | |
| 359 | 547 | NAPI |
| 360 | 548 | ---- |
| 549 | + NAPI (Rx polling mode) is supported in the e1000 driver. NAPI is enabled | |
| 550 | + or disabled based on the configuration of the kernel. To override | |
| 551 | + the default, use the following compile-time flags. | |
| 361 | 552 | |
| 362 | - NAPI (Rx polling mode) is supported in the e1000 driver. NAPI is enabled | |
| 363 | - or disabled based on the configuration of the kernel. | |
| 553 | + To enable NAPI, select the "Use Rx Polling (NAPI)" option when you enable | |
| 554 | + the e1000 driver | |
| 364 | 555 | |
| 365 | 556 | See www.cyberus.ca/~hadi/usenix-paper.tgz for more information on NAPI. |
| 366 | 557 |
| @@ -368,14 +559,97 @@ Additional Configurations | ||
| 368 | 559 | Known Issues |
| 369 | 560 | ============ |
| 370 | 561 | |
| 371 | - Jumbo Frames System Requirement | |
| 372 | - ------------------------------- | |
| 373 | - | |
| 374 | - Memory allocation failures have been observed on Linux systems with 64 MB | |
| 375 | - of RAM or less that are running Jumbo Frames. If you are using Jumbo Frames, | |
| 376 | - your system may require more than the advertised minimum requirement of 64 MB | |
| 377 | - of system memory. | |
| 378 | - | |
| 562 | +Dropped Receive Packets on Half-duplex 10/100 Networks | |
| 563 | +------------------------------------------------------ | |
| 564 | +If you have an Intel PCI Express adapter running at 10mbps or 100mbps, half- | |
| 565 | +duplex, you may observe occasional dropped receive packets. There are no | |
| 566 | +workarounds for this problem in this network configuration. The network must | |
| 567 | +be updated to operate in full-duplex, and/or 1000mbps only. | |
| 568 | + | |
| 569 | +Jumbo Frames System Requirement | |
| 570 | +------------------------------- | |
| 571 | +Memory allocation failures have been observed on Linux systems with 64 MB | |
| 572 | +of RAM or less that are running Jumbo Frames. If you are using Jumbo | |
| 573 | +Frames, your system may require more than the advertised minimum | |
| 574 | +requirement of 64 MB of system memory. | |
| 575 | + | |
| 576 | +Performance Degradation with Jumbo Frames | |
| 577 | +----------------------------------------- | |
| 578 | +Degradation in throughput performance may be observed in some Jumbo frames | |
| 579 | +environments. If this is observed, increasing the application's socket | |
| 580 | +buffer size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values | |
| 581 | +may help. See the specific application manual and | |
| 582 | +/usr/src/linux*/Documentation/ networking/ip-sysctl.txt for more details. | |
| 583 | + | |
| 584 | +Jumbo Frames on Foundry BigIron 8000 switch | |
| 585 | +------------------------------------------- | |
| 586 | +There is a known issue using Jumbo frames when connected to a Foundry | |
| 587 | +BigIron 8000 switch. This is a 3rd party limitation. If you experience | |
| 588 | +loss of packets, lower the MTU size. | |
| 589 | + | |
| 590 | +Allocating Rx Buffers when Using Jumbo Frames | |
| 591 | +--------------------------------------------- | |
| 592 | +Allocating Rx buffers when using Jumbo Frames on 2.6.x kernels may fail if | |
| 593 | +the available memory is heavily fragmented. This issue may be seen with PCI-X | |
| 594 | +adapters or with packet split disabled. This can be reduced or eliminated | |
| 595 | +by changing the amount of available memory for receive buffer allocation, by | |
| 596 | +increasing /proc/sys/vm/min_free_kbytes. | |
| 597 | + | |
| 598 | +Multiple Interfaces on Same Ethernet Broadcast Network | |
| 599 | +------------------------------------------------------ | |
| 600 | +Due to the default ARP behavior on Linux, it is not possible to have | |
| 601 | +one system on two IP networks in the same Ethernet broadcast domain | |
| 602 | +(non-partitioned switch) behave as expected. All Ethernet interfaces | |
| 603 | +will respond to IP traffic for any IP address assigned to the system. | |
| 604 | +This results in unbalanced receive traffic. | |
| 605 | + | |
| 606 | +If you have multiple interfaces in a server, either turn on ARP | |
| 607 | +filtering by entering: | |
| 608 | + | |
| 609 | + echo 1 > /proc/sys/net/ipv4/conf/all/arp_filter | |
| 610 | +(this only works if your kernel's version is higher than 2.4.5), | |
| 611 | + | |
| 612 | +NOTE: This setting is not saved across reboots. The configuration | |
| 613 | +change can be made permanent by adding the line: | |
| 614 | + net.ipv4.conf.all.arp_filter = 1 | |
| 615 | +to the file /etc/sysctl.conf | |
| 616 | + | |
| 617 | + or, | |
| 618 | + | |
| 619 | +install the interfaces in separate broadcast domains (either in | |
| 620 | +different switches or in a switch partitioned to VLANs). | |
| 621 | + | |
| 622 | +82541/82547 can't link or are slow to link with some link partners | |
| 623 | +----------------------------------------------------------------- | |
| 624 | +There is a known compatibility issue with 82541/82547 and some | |
| 625 | +low-end switches where the link will not be established, or will | |
| 626 | +be slow to establish. In particular, these switches are known to | |
| 627 | +be incompatible with 82541/82547: | |
| 628 | + | |
| 629 | + Planex FXG-08TE | |
| 630 | + I-O Data ETG-SH8 | |
| 631 | + | |
| 632 | +To workaround this issue, the driver can be compiled with an override | |
| 633 | +of the PHY's master/slave setting. Forcing master or forcing slave | |
| 634 | +mode will improve time-to-link. | |
| 635 | + | |
| 636 | + # make CFLAGS_EXTRA=-DE1000_MASTER_SLAVE=<n> | |
| 637 | + | |
| 638 | +Where <n> is: | |
| 639 | + | |
| 640 | + 0 = Hardware default | |
| 641 | + 1 = Master mode | |
| 642 | + 2 = Slave mode | |
| 643 | + 3 = Auto master/slave | |
| 644 | + | |
| 645 | +Disable rx flow control with ethtool | |
| 646 | +------------------------------------ | |
| 647 | +In order to disable receive flow control using ethtool, you must turn | |
| 648 | +off auto-negotiation on the same command line. | |
| 649 | + | |
| 650 | +For example: | |
| 651 | + | |
| 652 | + ethtool -A eth? autoneg off rx off | |
| 379 | 653 | |
| 380 | 654 | Support |
| 381 | 655 | ======= |
| @@ -384,20 +658,45 @@ For general information, go to the Intel support website at: | ||
| 384 | 658 | |
| 385 | 659 | http://support.intel.com |
| 386 | 660 | |
| 661 | +or the Intel Wired Networking project hosted by Sourceforge at: | |
| 662 | + | |
| 663 | + http://sourceforge.net/projects/e1000 | |
| 664 | + | |
| 387 | 665 | If an issue is identified with the released source code on the supported |
| 388 | -kernel with a supported adapter, email the specific information related to | |
| 389 | -the issue to linux.nics@intel.com. | |
| 666 | +kernel with a supported adapter, email the specific information related | |
| 667 | +to the issue to e1000-devel@lists.sf.net | |
| 668 | + | |
| 390 | 669 | |
| 391 | 670 | |
| 392 | 671 | License |
| 393 | 672 | ======= |
| 394 | 673 | |
| 395 | -This software program is released under the terms of a license agreement | |
| 396 | -between you ('Licensee') and Intel. Do not use or load this software or any | |
| 397 | -associated materials (collectively, the 'Software') until you have carefully | |
| 398 | -read the full terms and conditions of the LICENSE located in this software | |
| 399 | -package. By loading or using the Software, you agree to the terms of this | |
| 400 | -Agreement. If you do not agree with the terms of this Agreement, do not | |
| 401 | -install or use the Software. | |
| 674 | +Intel PRO/1000 Linux driver. | |
| 675 | +Copyright(c) 1999 - 2006 Intel Corporation. | |
| 676 | + | |
| 677 | +This program is free software; you can redistribute it and/or modify it | |
| 678 | +under the terms and conditions of the GNU General Public License, | |
| 679 | +version 2, as published by the Free Software Foundation. | |
| 680 | + | |
| 681 | +This program is distributed in the hope it will be useful, but WITHOUT | |
| 682 | +ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or | |
| 683 | +FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for | |
| 684 | +more details. | |
| 685 | + | |
| 686 | +You should have received a copy of the GNU General Public License along with | |
| 687 | +this program; if not, write to the Free Software Foundation, Inc., | |
| 688 | +51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. | |
| 689 | + | |
| 690 | +The full GNU General Public License is included in this distribution in | |
| 691 | +the file called "COPYING". | |
| 692 | + | |
| 693 | + | |
| 694 | + | |
| 695 | +Trademarks | |
| 696 | +========== | |
| 697 | + | |
| 698 | +Intel, Itanium, and Pentium are trademarks or registered trademarks of | |
| 699 | +Intel Corporation or its subsidiaries in the United States and other | |
| 700 | +countries. | |
| 402 | 701 | |
| 403 | 702 | * Other names and brands may be claimed as the property of others. |
Linux Kernel Documents
Fork