1 ============ 1 ============
2 Introduction 2 Introduction
3 ============ 3 ============
4 4
5 The RapidIO standard is a packet-based fabric 5 The RapidIO standard is a packet-based fabric interconnect standard designed for
6 use in embedded systems. Development of the Ra 6 use in embedded systems. Development of the RapidIO standard is directed by the
7 RapidIO Trade Association (RTA). The current v 7 RapidIO Trade Association (RTA). The current version of the RapidIO specification
8 is publicly available for download from the RT 8 is publicly available for download from the RTA web-site [1].
9 9
10 This document describes the basics of the Linu 10 This document describes the basics of the Linux RapidIO subsystem and provides
11 information on its major components. 11 information on its major components.
12 12
13 1 Overview 13 1 Overview
14 ========== 14 ==========
15 15
16 Because the RapidIO subsystem follows the Linu 16 Because the RapidIO subsystem follows the Linux device model it is integrated
17 into the kernel similarly to other buses by de 17 into the kernel similarly to other buses by defining RapidIO-specific device and
18 bus types and registering them within the devi 18 bus types and registering them within the device model.
19 19
20 The Linux RapidIO subsystem is architecture in 20 The Linux RapidIO subsystem is architecture independent and therefore defines
21 architecture-specific interfaces that provide 21 architecture-specific interfaces that provide support for common RapidIO
22 subsystem operations. 22 subsystem operations.
23 23
24 2. Core Components 24 2. Core Components
25 ================== 25 ==================
26 26
27 A typical RapidIO network is a combination of 27 A typical RapidIO network is a combination of endpoints and switches.
28 Each of these components is represented in the 28 Each of these components is represented in the subsystem by an associated data
29 structure. The core logical components of the 29 structure. The core logical components of the RapidIO subsystem are defined
30 in include/linux/rio.h file. 30 in include/linux/rio.h file.
31 31
32 2.1 Master Port 32 2.1 Master Port
33 --------------- 33 ---------------
34 34
35 A master port (or mport) is a RapidIO interfac 35 A master port (or mport) is a RapidIO interface controller that is local to the
36 processor executing the Linux code. A master p 36 processor executing the Linux code. A master port generates and receives RapidIO
37 packets (transactions). In the RapidIO subsyst 37 packets (transactions). In the RapidIO subsystem each master port is represented
38 by a rio_mport data structure. This structure 38 by a rio_mport data structure. This structure contains master port specific
39 resources such as mailboxes and doorbells. The 39 resources such as mailboxes and doorbells. The rio_mport also includes a unique
40 host device ID that is valid when a master por 40 host device ID that is valid when a master port is configured as an enumerating
41 host. 41 host.
42 42
43 RapidIO master ports are serviced by subsystem 43 RapidIO master ports are serviced by subsystem specific mport device drivers
44 that provide functionality defined for this su 44 that provide functionality defined for this subsystem. To provide a hardware
45 independent interface for RapidIO subsystem op 45 independent interface for RapidIO subsystem operations, rio_mport structure
46 includes rio_ops data structure which contains 46 includes rio_ops data structure which contains pointers to hardware specific
47 implementations of RapidIO functions. 47 implementations of RapidIO functions.
48 48
49 2.2 Device 49 2.2 Device
50 ---------- 50 ----------
51 51
52 A RapidIO device is any endpoint (other than m 52 A RapidIO device is any endpoint (other than mport) or switch in the network.
53 All devices are presented in the RapidIO subsy 53 All devices are presented in the RapidIO subsystem by corresponding rio_dev data
54 structure. Devices form one global device list 54 structure. Devices form one global device list and per-network device lists
55 (depending on number of available mports and n 55 (depending on number of available mports and networks).
56 56
57 2.3 Switch 57 2.3 Switch
58 ---------- 58 ----------
59 59
60 A RapidIO switch is a special class of device 60 A RapidIO switch is a special class of device that routes packets between its
61 ports towards their final destination. The pac 61 ports towards their final destination. The packet destination port within a
62 switch is defined by an internal routing table 62 switch is defined by an internal routing table. A switch is presented in the
63 RapidIO subsystem by rio_dev data structure ex 63 RapidIO subsystem by rio_dev data structure expanded by additional rio_switch
64 data structure, which contains switch specific 64 data structure, which contains switch specific information such as copy of the
65 routing table and pointers to switch specific 65 routing table and pointers to switch specific functions.
66 66
67 The RapidIO subsystem defines the format and i 67 The RapidIO subsystem defines the format and initialization method for subsystem
68 specific switch drivers that are designed to p 68 specific switch drivers that are designed to provide hardware-specific
69 implementation of common switch management rou 69 implementation of common switch management routines.
70 70
71 2.4 Network 71 2.4 Network
72 ----------- 72 -----------
73 73
74 A RapidIO network is a combination of intercon 74 A RapidIO network is a combination of interconnected endpoint and switch devices.
75 Each RapidIO network known to the system is re 75 Each RapidIO network known to the system is represented by corresponding rio_net
76 data structure. This structure includes lists 76 data structure. This structure includes lists of all devices and local master
77 ports that form the same network. It also cont 77 ports that form the same network. It also contains a pointer to the default
78 master port that is used to communicate with d 78 master port that is used to communicate with devices within the network.
79 79
80 2.5 Device Drivers 80 2.5 Device Drivers
81 ------------------ 81 ------------------
82 82
83 RapidIO device-specific drivers follow Linux K 83 RapidIO device-specific drivers follow Linux Kernel Driver Model and are
84 intended to support specific RapidIO devices a 84 intended to support specific RapidIO devices attached to the RapidIO network.
85 85
86 2.6 Subsystem Interfaces 86 2.6 Subsystem Interfaces
87 ------------------------ 87 ------------------------
88 88
89 RapidIO interconnect specification defines fea 89 RapidIO interconnect specification defines features that may be used to provide
90 one or more common service layers for all part 90 one or more common service layers for all participating RapidIO devices. These
91 common services may act separately from device 91 common services may act separately from device-specific drivers or be used by
92 device-specific drivers. Example of such servi 92 device-specific drivers. Example of such service provider is the RIONET driver
93 which implements Ethernet-over-RapidIO interfa 93 which implements Ethernet-over-RapidIO interface. Because only one driver can be
94 registered for a device, all common RapidIO se 94 registered for a device, all common RapidIO services have to be registered as
95 subsystem interfaces. This allows to have mult 95 subsystem interfaces. This allows to have multiple common services attached to
96 the same device without blocking attachment of 96 the same device without blocking attachment of a device-specific driver.
97 97
98 3. Subsystem Initialization 98 3. Subsystem Initialization
99 =========================== 99 ===========================
100 100
101 In order to initialize the RapidIO subsystem, 101 In order to initialize the RapidIO subsystem, a platform must initialize and
102 register at least one master port within the R 102 register at least one master port within the RapidIO network. To register mport
103 within the subsystem controller driver's initi 103 within the subsystem controller driver's initialization code calls function
104 rio_register_mport() for each available master 104 rio_register_mport() for each available master port.
105 105
106 After all active master ports are registered w 106 After all active master ports are registered with a RapidIO subsystem,
107 an enumeration and/or discovery routine may be 107 an enumeration and/or discovery routine may be called automatically or
108 by user-space command. 108 by user-space command.
109 109
110 RapidIO subsystem can be configured to be buil 110 RapidIO subsystem can be configured to be built as a statically linked or
111 modular component of the kernel (see details b 111 modular component of the kernel (see details below).
112 112
113 4. Enumeration and Discovery 113 4. Enumeration and Discovery
114 ============================ 114 ============================
115 115
116 4.1 Overview 116 4.1 Overview
117 ------------ 117 ------------
118 118
119 RapidIO subsystem configuration options allow 119 RapidIO subsystem configuration options allow users to build enumeration and
120 discovery methods as statically linked compone 120 discovery methods as statically linked components or loadable modules.
121 An enumeration/discovery method implementation 121 An enumeration/discovery method implementation and available input parameters
122 define how any given method can be attached to 122 define how any given method can be attached to available RapidIO mports:
123 simply to all available mports OR individually 123 simply to all available mports OR individually to the specified mport device.
124 124
125 Depending on selected enumeration/discovery bu 125 Depending on selected enumeration/discovery build configuration, there are
126 several methods to initiate an enumeration and 126 several methods to initiate an enumeration and/or discovery process:
127 127
128 (a) Statically linked enumeration and discov 128 (a) Statically linked enumeration and discovery process can be started
129 automatically during kernel initialization t 129 automatically during kernel initialization time using corresponding module
130 parameters. This was the original method use 130 parameters. This was the original method used since introduction of RapidIO
131 subsystem. Now this method relies on enumera 131 subsystem. Now this method relies on enumerator module parameter which is
132 'rio-scan.scan' for existing basic enumerati 132 'rio-scan.scan' for existing basic enumeration/discovery method.
133 When automatic start of enumeration/discover 133 When automatic start of enumeration/discovery is used a user has to ensure
134 that all discovering endpoints are started b 134 that all discovering endpoints are started before the enumerating endpoint
135 and are waiting for enumeration to be comple 135 and are waiting for enumeration to be completed.
136 Configuration option CONFIG_RAPIDIO_DISC_TIM 136 Configuration option CONFIG_RAPIDIO_DISC_TIMEOUT defines time that discovering
137 endpoint waits for enumeration to be complet 137 endpoint waits for enumeration to be completed. If the specified timeout
138 expires the discovery process is terminated 138 expires the discovery process is terminated without obtaining RapidIO network
139 information. NOTE: a timed out discovery pro 139 information. NOTE: a timed out discovery process may be restarted later using
140 a user-space command as it is described belo 140 a user-space command as it is described below (if the given endpoint was
141 enumerated successfully). 141 enumerated successfully).
142 142
143 (b) Statically linked enumeration and discov 143 (b) Statically linked enumeration and discovery process can be started by
144 a command from user space. This initiation m 144 a command from user space. This initiation method provides more flexibility
145 for a system startup compared to the option 145 for a system startup compared to the option (a) above. After all participating
146 endpoints have been successfully booted, an 146 endpoints have been successfully booted, an enumeration process shall be
147 started first by issuing a user-space comman 147 started first by issuing a user-space command, after an enumeration is
148 completed a discovery process can be started 148 completed a discovery process can be started on all remaining endpoints.
149 149
150 (c) Modular enumeration and discovery proces 150 (c) Modular enumeration and discovery process can be started by a command from
151 user space. After an enumeration/discovery m 151 user space. After an enumeration/discovery module is loaded, a network scan
152 process can be started by issuing a user-spa 152 process can be started by issuing a user-space command.
153 Similar to the option (b) above, an enumerat 153 Similar to the option (b) above, an enumerator has to be started first.
154 154
155 (d) Modular enumeration and discovery proces 155 (d) Modular enumeration and discovery process can be started by a module
156 initialization routine. In this case an enum 156 initialization routine. In this case an enumerating module shall be loaded
157 first. 157 first.
158 158
159 When a network scan process is started it call 159 When a network scan process is started it calls an enumeration or discovery
160 routine depending on the configured role of a 160 routine depending on the configured role of a master port: host or agent.
161 161
162 Enumeration is performed by a master port if i 162 Enumeration is performed by a master port if it is configured as a host port by
163 assigning a host destination ID greater than o 163 assigning a host destination ID greater than or equal to zero. The host
164 destination ID can be assigned to a master por 164 destination ID can be assigned to a master port using various methods depending
165 on RapidIO subsystem build configuration: 165 on RapidIO subsystem build configuration:
166 166
167 (a) For a statically linked RapidIO subsyste 167 (a) For a statically linked RapidIO subsystem core use command line parameter
168 "rapidio.hdid=" with a list of destination I 168 "rapidio.hdid=" with a list of destination ID assignments in order of mport
169 device registration. For example, in a syste 169 device registration. For example, in a system with two RapidIO controllers
170 the command line parameter "rapidio.hdid=-1, 170 the command line parameter "rapidio.hdid=-1,7" will result in assignment of
171 the host destination ID=7 to the second Rapi 171 the host destination ID=7 to the second RapidIO controller, while the first
172 one will be assigned destination ID=-1. 172 one will be assigned destination ID=-1.
173 173
174 (b) If the RapidIO subsystem core is built a 174 (b) If the RapidIO subsystem core is built as a loadable module, in addition
175 to the method shown above, the host destinat 175 to the method shown above, the host destination ID(s) can be specified using
176 traditional methods of passing module parame 176 traditional methods of passing module parameter "hdid=" during its loading:
177 177
178 - from command line: "modprobe rapidio hdid= 178 - from command line: "modprobe rapidio hdid=-1,7", or
179 - from modprobe configuration file using con 179 - from modprobe configuration file using configuration command "options",
180 like in this example: "options rapidio hdi 180 like in this example: "options rapidio hdid=-1,7". An example of modprobe
181 configuration file is provided in the sect 181 configuration file is provided in the section below.
182 182
183 NOTES: 183 NOTES:
184 (i) if "hdid=" parameter is omitted all avai 184 (i) if "hdid=" parameter is omitted all available mport will be assigned
185 destination ID = -1; 185 destination ID = -1;
186 186
187 (ii) the "hdid=" parameter in systems with m 187 (ii) the "hdid=" parameter in systems with multiple mports can have
188 destination ID assignments omitted from the 188 destination ID assignments omitted from the end of list (default = -1).
189 189
190 If the host device ID for a specific master po 190 If the host device ID for a specific master port is set to -1, the discovery
191 process will be performed for it. 191 process will be performed for it.
192 192
193 The enumeration and discovery routines use Rap 193 The enumeration and discovery routines use RapidIO maintenance transactions
194 to access the configuration space of devices. 194 to access the configuration space of devices.
195 195
196 NOTE: If RapidIO switch-specific device driver 196 NOTE: If RapidIO switch-specific device drivers are built as loadable modules
197 they must be loaded before enumeration/discove 197 they must be loaded before enumeration/discovery process starts.
198 This requirement is cased by the fact that enu 198 This requirement is cased by the fact that enumeration/discovery methods invoke
199 vendor-specific callbacks on early stages. 199 vendor-specific callbacks on early stages.
200 200
201 4.2 Automatic Start of Enumeration and Discove 201 4.2 Automatic Start of Enumeration and Discovery
202 ---------------------------------------------- 202 ------------------------------------------------
203 203
204 Automatic enumeration/discovery start method i 204 Automatic enumeration/discovery start method is applicable only to built-in
205 enumeration/discovery RapidIO configuration se 205 enumeration/discovery RapidIO configuration selection. To enable automatic
206 enumeration/discovery start by existing basic 206 enumeration/discovery start by existing basic enumerator method set use boot
207 command line parameter "rio-scan.scan=1". 207 command line parameter "rio-scan.scan=1".
208 208
209 This configuration requires synchronized start 209 This configuration requires synchronized start of all RapidIO endpoints that
210 form a network which will be enumerated/discov 210 form a network which will be enumerated/discovered. Discovering endpoints have
211 to be started before an enumeration starts to 211 to be started before an enumeration starts to ensure that all RapidIO
212 controllers have been initialized and are read 212 controllers have been initialized and are ready to be discovered. Configuration
213 parameter CONFIG_RAPIDIO_DISC_TIMEOUT defines 213 parameter CONFIG_RAPIDIO_DISC_TIMEOUT defines time (in seconds) which
214 a discovering endpoint will wait for enumerati 214 a discovering endpoint will wait for enumeration to be completed.
215 215
216 When automatic enumeration/discovery start is 216 When automatic enumeration/discovery start is selected, basic method's
217 initialization routine calls rio_init_mports() 217 initialization routine calls rio_init_mports() to perform enumeration or
218 discovery for all known mport devices. 218 discovery for all known mport devices.
219 219
220 Depending on RapidIO network size and configur 220 Depending on RapidIO network size and configuration this automatic
221 enumeration/discovery start method may be diff 221 enumeration/discovery start method may be difficult to use due to the
222 requirement for synchronized start of all endp 222 requirement for synchronized start of all endpoints.
223 223
224 4.3 User-space Start of Enumeration and Discov 224 4.3 User-space Start of Enumeration and Discovery
225 ---------------------------------------------- 225 -------------------------------------------------
226 226
227 User-space start of enumeration and discovery 227 User-space start of enumeration and discovery can be used with built-in and
228 modular build configurations. For user-space c 228 modular build configurations. For user-space controlled start RapidIO subsystem
229 creates the sysfs write-only attribute file '/ 229 creates the sysfs write-only attribute file '/sys/bus/rapidio/scan'. To initiate
230 an enumeration or discovery process on specifi 230 an enumeration or discovery process on specific mport device, a user needs to
231 write mport_ID (not RapidIO destination ID) in 231 write mport_ID (not RapidIO destination ID) into that file. The mport_ID is a
232 sequential number (0 ... RIO_MAX_MPORTS) assig 232 sequential number (0 ... RIO_MAX_MPORTS) assigned during mport device
233 registration. For example for machine with sin 233 registration. For example for machine with single RapidIO controller, mport_ID
234 for that controller always will be 0. 234 for that controller always will be 0.
235 235
236 To initiate RapidIO enumeration/discovery on a 236 To initiate RapidIO enumeration/discovery on all available mports a user may
237 write '-1' (or RIO_MPORT_ANY) into the scan at 237 write '-1' (or RIO_MPORT_ANY) into the scan attribute file.
238 238
239 4.4 Basic Enumeration Method 239 4.4 Basic Enumeration Method
240 ---------------------------- 240 ----------------------------
241 241
242 This is an original enumeration/discovery meth 242 This is an original enumeration/discovery method which is available since
243 first release of RapidIO subsystem code. The e 243 first release of RapidIO subsystem code. The enumeration process is
244 implemented according to the enumeration algor 244 implemented according to the enumeration algorithm outlined in the RapidIO
245 Interconnect Specification: Annex I [1]. 245 Interconnect Specification: Annex I [1].
246 246
247 This method can be configured as statically li 247 This method can be configured as statically linked or loadable module.
248 The method's single parameter "scan" allows to 248 The method's single parameter "scan" allows to trigger the enumeration/discovery
249 process from module initialization routine. 249 process from module initialization routine.
250 250
251 This enumeration/discovery method can be start 251 This enumeration/discovery method can be started only once and does not support
252 unloading if it is built as a module. 252 unloading if it is built as a module.
253 253
254 The enumeration process traverses the network 254 The enumeration process traverses the network using a recursive depth-first
255 algorithm. When a new device is found, the enu 255 algorithm. When a new device is found, the enumerator takes ownership of that
256 device by writing into the Host Device ID Lock 256 device by writing into the Host Device ID Lock CSR. It does this to ensure that
257 the enumerator has exclusive right to enumerat 257 the enumerator has exclusive right to enumerate the device. If device ownership
258 is successfully acquired, the enumerator alloc 258 is successfully acquired, the enumerator allocates a new rio_dev structure and
259 initializes it according to device capabilitie 259 initializes it according to device capabilities.
260 260
261 If the device is an endpoint, a unique device 261 If the device is an endpoint, a unique device ID is assigned to it and its value
262 is written into the device's Base Device ID CS 262 is written into the device's Base Device ID CSR.
263 263
264 If the device is a switch, the enumerator allo 264 If the device is a switch, the enumerator allocates an additional rio_switch
265 structure to store switch specific information 265 structure to store switch specific information. Then the switch's vendor ID and
266 device ID are queried against a table of known 266 device ID are queried against a table of known RapidIO switches. Each switch
267 table entry contains a pointer to a switch-spe 267 table entry contains a pointer to a switch-specific initialization routine that
268 initializes pointers to the rest of switch spe 268 initializes pointers to the rest of switch specific operations, and performs
269 hardware initialization if necessary. A RapidI 269 hardware initialization if necessary. A RapidIO switch does not have a unique
270 device ID; it relies on hopcount and routing f 270 device ID; it relies on hopcount and routing for device ID of an attached
271 endpoint if access to its configuration regist 271 endpoint if access to its configuration registers is required. If a switch (or
272 chain of switches) does not have any endpoint 272 chain of switches) does not have any endpoint (except enumerator) attached to
273 it, a fake device ID will be assigned to confi 273 it, a fake device ID will be assigned to configure a route to that switch.
274 In the case of a chain of switches without end 274 In the case of a chain of switches without endpoint, one fake device ID is used
275 to configure a route through the entire chain 275 to configure a route through the entire chain and switches are differentiated by
276 their hopcount value. 276 their hopcount value.
277 277
278 For both endpoints and switches the enumerator 278 For both endpoints and switches the enumerator writes a unique component tag
279 into device's Component Tag CSR. That unique v 279 into device's Component Tag CSR. That unique value is used by the error
280 management notification mechanism to identify 280 management notification mechanism to identify a device that is reporting an
281 error management event. 281 error management event.
282 282
283 Enumeration beyond a switch is completed by it 283 Enumeration beyond a switch is completed by iterating over each active egress
284 port of that switch. For each active link, a r 284 port of that switch. For each active link, a route to a default device ID
285 (0xFF for 8-bit systems and 0xFFFF for 16-bit 285 (0xFF for 8-bit systems and 0xFFFF for 16-bit systems) is temporarily written
286 into the routing table. The algorithm recurs b 286 into the routing table. The algorithm recurs by calling itself with hopcount + 1
287 and the default device ID in order to access t 287 and the default device ID in order to access the device on the active port.
288 288
289 After the host has completed enumeration of th 289 After the host has completed enumeration of the entire network it releases
290 devices by clearing device ID locks (calls rio 290 devices by clearing device ID locks (calls rio_clear_locks()). For each endpoint
291 in the system, it sets the Discovered bit in t 291 in the system, it sets the Discovered bit in the Port General Control CSR
292 to indicate that enumeration is completed and 292 to indicate that enumeration is completed and agents are allowed to execute
293 passive discovery of the network. 293 passive discovery of the network.
294 294
295 The discovery process is performed by agents a 295 The discovery process is performed by agents and is similar to the enumeration
296 process that is described above. However, the 296 process that is described above. However, the discovery process is performed
297 without changes to the existing routing becaus 297 without changes to the existing routing because agents only gather information
298 about RapidIO network structure and are buildi 298 about RapidIO network structure and are building an internal map of discovered
299 devices. This way each Linux-based component o 299 devices. This way each Linux-based component of the RapidIO subsystem has
300 a complete view of the network. The discovery 300 a complete view of the network. The discovery process can be performed
301 simultaneously by several agents. After initia 301 simultaneously by several agents. After initializing its RapidIO master port
302 each agent waits for enumeration completion by 302 each agent waits for enumeration completion by the host for the configured wait
303 time period. If this wait time period expires 303 time period. If this wait time period expires before enumeration is completed,
304 an agent skips RapidIO discovery and continues 304 an agent skips RapidIO discovery and continues with remaining kernel
305 initialization. 305 initialization.
306 306
307 4.5 Adding New Enumeration/Discovery Method 307 4.5 Adding New Enumeration/Discovery Method
308 ------------------------------------------- 308 -------------------------------------------
309 309
310 RapidIO subsystem code organization allows add 310 RapidIO subsystem code organization allows addition of new enumeration/discovery
311 methods as new configuration options without s 311 methods as new configuration options without significant impact to the core
312 RapidIO code. 312 RapidIO code.
313 313
314 A new enumeration/discovery method has to be a 314 A new enumeration/discovery method has to be attached to one or more mport
315 devices before an enumeration/discovery proces 315 devices before an enumeration/discovery process can be started. Normally,
316 method's module initialization routine calls r 316 method's module initialization routine calls rio_register_scan() to attach
317 an enumerator to a specified mport device (or 317 an enumerator to a specified mport device (or devices). The basic enumerator
318 implementation demonstrates this process. 318 implementation demonstrates this process.
319 319
320 4.6 Using Loadable RapidIO Switch Drivers 320 4.6 Using Loadable RapidIO Switch Drivers
321 ----------------------------------------- 321 -----------------------------------------
322 322
323 In the case when RapidIO switch drivers are bu 323 In the case when RapidIO switch drivers are built as loadable modules a user
324 must ensure that they are loaded before the en 324 must ensure that they are loaded before the enumeration/discovery starts.
325 This process can be automated by specifying pr 325 This process can be automated by specifying pre- or post- dependencies in the
326 RapidIO-specific modprobe configuration file a 326 RapidIO-specific modprobe configuration file as shown in the example below.
327 327
328 File /etc/modprobe.d/rapidio.conf:: 328 File /etc/modprobe.d/rapidio.conf::
329 329
330 # Configure RapidIO subsystem modules 330 # Configure RapidIO subsystem modules
331 331
332 # Set enumerator host destination ID (overri 332 # Set enumerator host destination ID (overrides kernel command line option)
333 options rapidio hdid=-1,2 333 options rapidio hdid=-1,2
334 334
335 # Load RapidIO switch drivers immediately af 335 # Load RapidIO switch drivers immediately after rapidio core module was loaded
336 softdep rapidio post: idt_gen2 idtcps tsi57x 336 softdep rapidio post: idt_gen2 idtcps tsi57x
337 337
338 # OR : 338 # OR :
339 339
340 # Load RapidIO switch drivers just before ri 340 # Load RapidIO switch drivers just before rio-scan enumerator module is loaded
341 softdep rio-scan pre: idt_gen2 idtcps tsi57x 341 softdep rio-scan pre: idt_gen2 idtcps tsi57x
342 342
343 -------------------------- 343 --------------------------
344 344
345 NOTE: 345 NOTE:
346 In the example above, one of "softdep" comma 346 In the example above, one of "softdep" commands must be removed or
347 commented out to keep required module loadin 347 commented out to keep required module loading sequence.
348 348
349 5. References 349 5. References
350 ============= 350 =============
351 351
352 [1] RapidIO Trade Association. RapidIO Interco 352 [1] RapidIO Trade Association. RapidIO Interconnect Specifications.
353 http://www.rapidio.org. 353 http://www.rapidio.org.
354 354
355 [2] Rapidio TA. Technology Comparisons. 355 [2] Rapidio TA. Technology Comparisons.
356 http://www.rapidio.org/education/technolog 356 http://www.rapidio.org/education/technology_comparisons/
357 357
358 [3] RapidIO support for Linux. 358 [3] RapidIO support for Linux.
359 https://lwn.net/Articles/139118/ 359 https://lwn.net/Articles/139118/
360 360
361 [4] Matt Porter. RapidIO for Linux. Ottawa Lin 361 [4] Matt Porter. RapidIO for Linux. Ottawa Linux Symposium, 2005
362 https://www.kernel.org/doc/ols/2005/ols200 362 https://www.kernel.org/doc/ols/2005/ols2005v2-pages-43-56.pdf
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.