Input/Output Configuration Program User's Guide For ICP Iocp

IBM
IBM Z
Input/Output Configuration Program User's Guide for ICP
IOCP
SB10-7172-01
IBM
IBM Z
Input/Output Configuration Program User's Guide for ICP
IOCP
SB10-7172-01
Note:
Before using this information and the product it supports, read the information in “Safety” on
page xi, Appendix G, “Notices,” on page 299, and IBM Systems Environmental Notices and User
Guide, Z125-5823.
This edition, SB10-7172-01 applies to ICP IOCP Version 5 Release 4 Modification 0 (5.4.0) (batch and stand-alone
versions). This edition replaces SB10-7172-00.
There might be a newer version of this document in a PDF file available on Resource Link. Go to
http://www.ibm.com/servers/resourcelink and click Library on the navigation bar.
© Copyright IBM Corporation 2017, 2018.
US Government Users Restricted Rights – Use, duplication or disclosure restricted by GSA ADP Schedule Contract
with IBM Corp.
Contents
Figures . . . . . . . . . . . . . . vii Using a PCHID report to help plan your I/O
configuration . . . . . . . . . . . . . 17
Tables . . . . . . . . . . . . . . . ix Sample PCHID report . . . . . . . . . . 18
I/O configuration process . . . . . . . . . 19
Card-image input to IOCP . . . . . . . . 20
Safety . . . . . . . . . . . . . . . xi Console input to IOCP . . . . . . . . . 20
Safety notices . . . . . . . . . . . . . . xi I/O configuration data sets . . . . . . . . . 20
World trade safety information . . . . . . . xi IOCDSs and CPC modes . . . . . . . . . 21
Laser safety information . . . . . . . . . . xi I/O configuration considerations . . . . . . . 21
Laser compliance . . . . . . . . . . . xi Defining multiple logical channel subsystems . . 21
Control unit headers . . . . . . . . . . 22
About this publication . . . . . . . . xiii Shared ESCON CTC considerations . . . . . . 25
What is included in this user's guide . . . . . xiv Shared CTC unit address ranges . . . . . . 25
Revisions . . . . . . . . . . . . . . . xv Shared CTC unit addresses . . . . . . . . 26
Related information . . . . . . . . . . . xv Coupling facility support . . . . . . . . . . 27
OS/390. . . . . . . . . . . . . . . xv Coupling facility channels . . . . . . . . 27
z/OS and z/OS.e . . . . . . . . . . . xv Open Systems Adapter (OSA) feature support . . . 31
z/VM . . . . . . . . . . . . . . . xvi OSA support facility . . . . . . . . . . 32
z/VSE . . . . . . . . . . . . . . . xvi OSA channels. . . . . . . . . . . . . 32
3907 models . . . . . . . . . . . . . xvi OSA-ICC channels (TYPE=OSC) . . . . . . 32
3906 models . . . . . . . . . . . . . xvi OSA QDIO channels (TYPE=OSD) . . . . . . 32
2965 models . . . . . . . . . . . . . xvi OSA non-QDIO channels (TYPE=OSE) . . . . 33
2964 models. . . . . . . . . . . . . xvii OSA for zBX management network (TYPE=OSM) 33
2828 models. . . . . . . . . . . . . xvii OSA for NCP channels (TYPE=OSN) . . . . . 33
2827 models. . . . . . . . . . . . . xvii OSA for zBX data network (TYPE=OSX) . . . . 33
2818 models. . . . . . . . . . . . . xvii OSA channels for OSA-ICC (TYPE=OSC), QDIO
2817 models. . . . . . . . . . . . . xvii (TYPE=OSD), non-QDIO (TYPE=OSE), zBX
2098 models . . . . . . . . . . . . xviii management network (TYPE=OSM), NCP
2097 models . . . . . . . . . . . . xviii (TYPE=OSN), and zBX data network
2096 models . . . . . . . . . . . . xviii (TYPE=OSX) . . . . . . . . . . . . . 33
2094 models . . . . . . . . . . . . xviii OSA configuration considerations . . . . . . 33
2086 models . . . . . . . . . . . . xviii Internal queued direct communication
2084 models . . . . . . . . . . . . . xix (HiperSockets) support . . . . . . . . . . 36
Enterprise systems connection (ESCON) . . . xix Defining IQD channels . . . . . . . . . 37
Fibre connection (FICON) . . . . . . . . xix Fibre Channel Protocol support. . . . . . . . 37
Open Systems Adapter (OSA) . . . . . . . xix N_Port ID virtualization (NPIV) . . . . . . 38
Note to all users . . . . . . . . . . . . xix Defining FICON switch configurations . . . . . 38
Accessibility . . . . . . . . . . . . . . xix Managing logical paths for ESCON and FICON
Accessibility features . . . . . . . . . . xx channels . . . . . . . . . . . . . . . 40
Keyboard navigation . . . . . . . . . . xx Control unit allocation of logical paths . . . . 40
Consult assistive technologies . . . . . . . xx Why manage logical paths? . . . . . . . . 40
IBM and accessibility . . . . . . . . . . xx Managing the establishment of logical paths . . . 41
Related HMC and SE console information . . . . xx Logical path considerations . . . . . . . . 42
How to send your comments . . . . . . . . xx Recommendations for managing logical paths . . 43
Channel path selection . . . . . . . . . . 44
Summary of changes . . . . . . . . xxi Rotation order . . . . . . . . . . . . 44
Rotating path selection . . . . . . . . . 44
Path selection using a preferred path . . . . . 44
Chapter 1. IOCP introduction. . . . . . 1
Control unit types . . . . . . . . . . . . 45
Hardware I/O configuration . . . . . . . . . 1
Versions of IOCP . . . . . . . . . . . . 45
| 3907, 3906, 2965, 2964, 2828, 2827, 2818, and 2817
Overview of the z/OS version of IOCP . . . . 46
models . . . . . . . . . . . . . . . 1
IOCP and z/OS hardware configuration
2098 and 2097 models . . . . . . . . . . 8
definition . . . . . . . . . . . . . . 46
2096 and 2094 models . . . . . . . . . . 12
Overview of the VM version of IOCP. . . . . 47
2086 and 2084 models . . . . . . . . . . 14
IOCP and system generation of a VM operating
system . . . . . . . . . . . . . . . 47
© Copyright IBM Corp. 2017, 2018 iii

Overview of the z/VSE version of IOCP. . . . 48 JCL example of producing configuration reports 116
IOCP and system generation of a z/VSE Return codes . . . . . . . . . . . . . 116
operating system . . . . . . . . . . . 49 Methods of running the z/OS version of IOCP . . 117
Overview of the stand-alone version of IOCP . . 50 Initial definition of I/O configuration data . . 117
Redefinition and replacement of I/O
Chapter 2. IOCP statements and configuration data . . . . . . . . . . . 118
functions . . . . . . . . . . . . . . 53
Using this chapter . . . . . . . . . . . . 53 Chapter 4. Running the VM version of
Specifying IOCP statements . . . . . . . . . 53 IOCP . . . . . . . . . . . . . . . 119
IOCP and z/OS I/O configuration. . . . . . . 54 General information . . . . . . . . . . . 119
IOCP and z/VM I/O configuration . . . . . . 54 Hardware configuration definition (HCD) notes 119
IOCP and z/VSE I/O configuration . . . . . . 55 IOCP storage requirements . . . . . . . . . 119
IOCP definitions. . . . . . . . . . . . . 56 The IOCP command . . . . . . . . . . . 120
ID . . . . . . . . . . . . . . . . . 58 Program options . . . . . . . . . . . 121
ID Statement coding example . . . . . . . 59 IOCDS generation options . . . . . . . . 121
RESOURCE . . . . . . . . . . . . . . 60 Configuration report options . . . . . . . 123
RESOURCE statement coding examples . . . . . 63 Listing control options . . . . . . . . . 123
Example 1 . . . . . . . . . . . . . . 63 System terminal options . . . . . . . . . 124
Example 2 . . . . . . . . . . . . . . 63 IOCP CMS utility options . . . . . . . . 124
Example 3 . . . . . . . . . . . . . . 63 IOCP command – general usage notes . . . . 125
CHPID . . . . . . . . . . . . . . . . 63 IOCP command – usage notes for VM . . . . 125
CHPID statement coding examples . . . . . . 79 Examples of generating an IOCDS . . . . . 126
Example 1 . . . . . . . . . . . . . . 79 Examples of producing a configuration report 127
Example 2 . . . . . . . . . . . . . . 79 Return codes . . . . . . . . . . . . . 127
Example 3 . . . . . . . . . . . . . . 79 Methods of running the VM version of IOCP. . . 128
Example 4 . . . . . . . . . . . . . . 79 Initial definition of I/O configuration data . . 128
Example 5 . . . . . . . . . . . . . . 80 Redefinition and replacement of I/O
Example 6 . . . . . . . . . . . . . . 80 configuration data . . . . . . . . . . . 129
Example 7 . . . . . . . . . . . . . . 80
CNTLUNIT . . . . . . . . . . . . . . 80 Chapter 5. Running the z/VSE version
CNTLUNIT statement coding examples . . . . . 90 of IOCP . . . . . . . . . . . . . . 131
Example 1 . . . . . . . . . . . . . . 90
General information . . . . . . . . . . . 131
Example 2 . . . . . . . . . . . . . . 90
IOCP storage requirements . . . . . . . . . 131
Example 3 . . . . . . . . . . . . . . 90
Coding the JCL EXEC statement for generating an
Example 4 . . . . . . . . . . . . . . 91
IOCDS . . . . . . . . . . . . . . . 131
Example 5 . . . . . . . . . . . . . . 91
JCL examples of generating an IOCDS . . . . 133
Example 6 . . . . . . . . . . . . . . 91
Coding the JCL EXEC statement for producing
IODEVICE. . . . . . . . . . . . . . . 92
configuration reports . . . . . . . . . . . 135
IODEVICE statement coding examples . . . . 101
Conditions for coding the REPORT option. . . 135
UUID . . . . . . . . . . . . . . . . 102
JCL example of producing configuration reports 135
UUID statement coding examples . . . . . 103
Return codes . . . . . . . . . . . . . 136
FUNCTION . . . . . . . . . . . . . . 103
Methods of running the z/VSE version of IOCP 136
FUNCTION statement coding examples . . . 108
Initial definition of I/O configuration data . . 137
IOCP functions . . . . . . . . . . . . . 109
Redefinition and replacement of I/O
Generating an IOCDS . . . . . . . . . 109
configuration data . . . . . . . . . . . 137
Producing configuration reports . . . . . . 110
Chapter 6. Running the stand-alone

Chapter 3. Running the z/OS version
version of IOCP . . . . . . . . . . 139
of IOCP . . . . . . . . . . . . . . 111
Generating an input/output configuration data set 139
Hardware configuration definition (HCD) notes: 111
Creating an IOCDS source file for a new CPC 139
Authorizing an IOCDS write . . . . . . . . 112
Starting the program . . . . . . . . . . . 141
Building the IOCP module . . . . . . . . . 112
Input/Output configuration window . . . . . 141
IOCP storage requirements . . . . . . . . . 112
IOCP options . . . . . . . . . . . . . 142
Coding the JCL EXEC statement for generating an
Overview of options . . . . . . . . . . 143
IOCDS. . . . . . . . . . . . . . . . 112
IOCP view actions. . . . . . . . . . . . 144
JCL examples of generating an IOCDS . . . . 114
Methods of running the stand-alone version of
Coding the JCL EXEC statement for producing
IOCP . . . . . . . . . . . . . . . . 144
configuration reports . . . . . . . . . . . 115
Starter IOCDS information for the stand-alone
Conditions for coding the REPORT options . . 116
version of IOCP . . . . . . . . . . . 144
iv IOCP User's Guide

Creating an IOCDS in a new CPC . . . . . 145 ESCON CTC dynamic through an ESCON director
Updating an IOCDS using the card-image input with multiple paths . . . . . . . . . . . 255
editor . . . . . . . . . . . . . . . 146 IOCP input for the configuration . . . . . . 256
Description of a stand-alone build of an IOCDS 147 FICON CTC and director with single path. . . . 256
IOCP wait state codes . . . . . . . . . . 148 IOCP input for the configuration . . . . . . 257
Enabled wait states . . . . . . . . . . 148 Chained ESCON directors . . . . . . . . . 257
Disabled wait states . . . . . . . . . . 148 IOCP input for system 1 configuration . . . . 258
IOCP input for system 2 configuration . . . . 258
Chapter 7. IOCP configuration reports 151 Cascaded FICON directors . . . . . . . . . 258
Configuration report headings. . . . . . . . 151 IOCP input for the configuration . . . . . . 259
IOCDS totals report . . . . . . . . . . . 152 Control unit logical addressing . . . . . . . 259
Sample IOCDS totals report . . . . . . . 153 IOCP input for the configuration . . . . . . 260
Partition report . . . . . . . . . . . . . 153 Channels through an ESCON converter . . . . 260
Sample partition report . . . . . . . . . 154 IOCP input for the configuration . . . . . . 261
FID summary report . . . . . . . . . . . 155 FCV channel configuration . . . . . . . . . 261
Sample FID summary report . . . . . . . 156 IOCP input for the configuration . . . . . . 262
PCHID summary report . . . . . . . . . . 156 ESCON control units requiring unit address 00 . . 262
Sample PCHID summary report . . . . . . 159 OSA configuration with two logical partitions . . 262
Channel path identifier (CHPID) summary report 159 IOCP input for the configuration . . . . . . 263
Sample CHPID summary report . . . . . . 161 OSA-ICC configuration . . . . . . . . . . 263
I/O device report . . . . . . . . . . . . 162 IOCP input for the configuration . . . . . . 264
Sample I/O device report . . . . . . . . 165 Partial manual session configuration file . . . 264
Channel path report . . . . . . . . . . . 167 FCP channel configuration . . . . . . . . . 265
Sample channel path report . . . . . . . 169 Partial host operating system configuration
Conditions for producing reports . . . . . . . 170 definition. . . . . . . . . . . . . . 266
IOCP input for the configuration . . . . . . 266
Overview of the configuration definitions . . . 266
Chapter 8. IOCP messages . . . . . 171
Coupling facility configuration . . . . . . . 266
IOCP messages (ICP prefix) . . . . . . . . 171
IOCP input for z/OS configuration . . . . . 267
IOCP messages (DMSICP prefix) . . . . . . 233
IOCP input for coupling facility configuration 267
ESCON CTC with shared and unshared channels 267
Appendix A. Coding IOCP statements 237 IOCP input for System A (shared) . . . . . 268
Rules for coding IOCP statements . . . . . . 237 IOCP input for System B (unshared). . . . . 269
Format and coding conventions . . . . . . . 238 ESCON CTC with shared channels . . . . . . 271
Example . . . . . . . . . . . . . . 238 IOCP input for System A (shared) . . . . . 272
IOCP input for System B (unshared). . . . . 275
Appendix B. Lists of statement input 241 FICON CTC with shared channels . . . . . . 277
IOCP statements for System A. . . . . . . 278
Appendix C. Characteristics of the I/O IOCP statements for System B . . . . . . . 281
interface timeout function . . . . . . 243
Timeout considerations . . . . . . . . . . 244 Appendix F. Machine limits and rules 285
Machine rules . . . . . . . . . . . . . 285
Machine limits . . . . . . . . . . . . . 286
Appendix D. List of I/O devices and
Keyword applicability for channel paths and
control units . . . . . . . . . . . . 245 functions . . . . . . . . . . . . . . . 294
CPC activation and power-on reset error . . . . 296
Appendix E. Configuration examples 253
More than eight paths to an ESCON device . . . 253 Appendix G. Notices . . . . . . . . 299
IOCP input for the configuration . . . . . . 253 Trademarks . . . . . . . . . . . . . . 300
Shared paths to an ESCON device . . . . . . 254 Class A Notices. . . . . . . . . . . . . 300
IOCP input for the configuration . . . . . . 254
ESCON CTC point-to-point. . . . . . . . . 254
Index . . . . . . . . . . . . . . . 305
Contents v
vi IOCP User's Guide
Figures
1. I/O channel path configuration for a 22. Sample FID summary report . . . . . . 156
representative PCIe I/O drawer in a 2817, 2818, 23. Sample PCHID summary report . . . . . 159
2827, 2828, 2964, 2965, and 3906 model . . . . 2 24. Sample CHPID summary report (Part 1 of 2) 161
| 2. I/O channel path configuration for a 25. Sample CHPID summary report (Part 2 of 2) 162
| representative PCIe I/O drawer in a 3907 model 3 26. Sample I/O device report (Part 1 of 3) 165
3. I/O channel path configuration for a 27. Sample I/O device report (Part 2 of 3) 166
representative I/O cage in a 2827 and 2817 28. Sample I/O device report (Part 3 of 3) 167
model . . . . . . . . . . . . . . . 5 29. Sample channel path report (Part 1 of 2) 169
4. I/O channel path configuration for a 30. Sample channel path report (Part 2 of 2) 170
representative I/O drawer in a 2965, 2964, 2828, 31. Sample IOCP statements . . . . . . . . 241
2817, 2818, and 2827 model . . . . . . . 6 32. Sample SYSTEM CONFIG statements for VM
5. I/O channel path configuration for a I/O definition . . . . . . . . . . . 241
representative I/O cage in a 2097 model . . . 9 33. HCPRIO file for VM system generation 242
6. I/O channel path configuration for a 34. Source input for VSE system generation 242
representative I/O drawer in a 2098 model . . 10 35. LPAR mode with more than eight paths to an
7. I/O channel path configuration for a ESCON device . . . . . . . . . . . 253
representative I/O cage in a 2094 or 2096 36. LPAR mode with shared paths to an ESCON
model . . . . . . . . . . . . . . 12 device . . . . . . . . . . . . . . 254
8. I/O channel path configuration for a 37. ESCON CTC point-to-point . . . . . . . 255
representative I/O cage in a 2084 or 2086 38. ESCON CTC dynamic through an ESCON
model . . . . . . . . . . . . . . 15 director with multiple paths . . . . . . 256
9. Sample PCHID report (Part 1 of 2) . . . . . 18 39. FICON CTC and director with single path 257
10. Sample PCHID report (Part 2 of 2) . . . . . 19 40. Chained ESCON directors . . . . . . . 258
11. Valid control unit headers . . . . . . . . 23 41. Cascaded FICON directors . . . . . . . 259
12. Control unit header that is incorrect . . . . 25 42. Control unit logical addressing. . . . . . 260
13. Overview of z/OS version of IOCP . . . . 46 43. Channels through an ESCON converter 261
14. Overview of VM version of IOCP . . . . . 48 44. FCV channel configuration . . . . . . . 262
15. Overview of z/VSE version of IOCP . . . . 49 45. ESCON control units requiring unit address
16. Overview of stand-alone version of IOCP 51 00 . . . . . . . . . . . . . . . 262
17. IOCP statements . . . . . . . . . . . 54 46. OSA configuration with two logical partitions 263
18. Correspondence between IOCP statements and 47. OSA-ICC . . . . . . . . . . . . . 264
SYSTEM CONFIG statements or HCPRIO 48. FCP channel configuration . . . . . . . 265
statements . . . . . . . . . . . . . 55 49. Coupling facility configuration . . . . . . 266
19. Correspondence between IOCP statements and 50. ESCON CTC with shared and unshared
z/VSE IPL ADD statements . . . . . . . 56 channels . . . . . . . . . . . . . 268
20. Sample IOCDS totals report. . . . . . . 153 51. ESCON CTC with shared channels . . . . 272
21. Sample partition report . . . . . . . . 154 52. FICON CTC (fully redundant) . . . . . . 278
© Copyright IBM Corp. 2017, 2018 vii

viii IOCP User's Guide
Tables
1. EC levels that ICP IOCP supports . . . . . xiii 14. IOCP parameter values for IODEVICE and
| 2. Resource group and corresponding slots for CNTLUNIT statements . . . . . . . . 245
| 3907 model . . . . . . . . . . . . . 3 15. z/OS considerations: . . . . . . . . . 249
3. Resource group and corresponding slots for | 16. Machine rules . . . . . . . . . . . 285
3906 model . . . . . . . . . . . . . 3 17. Constant machine limits . . . . . . . . 286
4. Resource group and corresponding slots for | 18. Machine limits . . . . . . . . . . . 287
2964 and 2965 models . . . . . . . . . 4 | 19. Valid CHPID types. . . . . . . . . . 291
5. Relationship of book jack numbers to MBAs 16 | 20. Valid FUNCTION types . . . . . . . . 292
6. Input device by IOCP version . . . . . . 19 21. Maximums by CPC . . . . . . . . . 293
7. Summary of connectivity . . . . . . . . 34 22. FCP Channel Path Limits . . . . . . . 293
8. Logical path summary by control unit . . . 40 23. Keyword applicability for channel paths 294
9. Supported CHPARM values . . . . . . . 77 24. Keyword applicability for functions . . . . 295
10. Valid VF ranges . . . . . . . . . . . 106 | 25. Statement and keyword applicability for
11. Return codes for z/OS version of IOCP 116 | machine types . . . . . . . . . . . 295
12. VM version of IOCP . . . . . . . . . 127 26. Activation and POR error message
13. Return codes for z/VSE version of IOCP 136 hexadecimal values . . . . . . . . . 296
© Copyright IBM Corp. 2017, 2018 ix

x IOCP User's Guide
Safety
Safety notices
Safety notices may be printed throughout this guide. DANGER notices warn you of conditions or
procedures that can result in death or severe personal injury. CAUTION notices warn you of conditions
or procedures that can cause personal injury that is neither lethal nor extremely hazardous. Attention
notices warn you of conditions or procedures that can cause damage to machines, equipment, or
programs.
World trade safety information

Several countries require the safety information contained in product publications to be presented in their
translation. If this requirement applies to your country, a safety information booklet is included in the
publications package shipped with the product. The booklet contains the translated safety information
with references to the US English source. Before using a US English publication to install, operate, or
service this product, you must first become familiar with the related safety information in the Systems
Safety Notices, G229-9054. You should also refer to the booklet any time you do not clearly understand
any safety information in the US English publications.
Laser safety information

All IBM Z® (Z) and IBM LinuxONE™ (LinuxONE) models can use I/O cards such as FICON®, Open
Systems Adapter (OSA), InterSystem Channel-3 (ISC-3), zHyperLinkTM Express, or other I/O features
which are fiber optic based and utilize lasers (short wavelength or long wavelength lasers).
Laser compliance
All lasers are certified in the US to conform to the requirements of DHHS 21 CFR Subchapter J for Class
1 or Class 1M laser products. Outside the US, they are certified to be in compliance with IEC 60825 as a
Class 1 or Class 1M laser product. Consult the label on each part for laser certification numbers and
approval information.
CAUTION: Data processing environments can contain equipment transmitting on system links with
laser modules that operate at greater than Class 1 power levels. For this reason, never look into the
end of an optical fiber cable or open receptacle. (C027)
CAUTION: This product contains a Class 1M laser. Do not view directly with optical instruments.
(C028)
© Copyright IBM Corp. 2017, 2018 xi

xii IOCP User's Guide
About this publication
This user's guide describes the Input/Output Configuration Program (ICP IOCP) that supports multiple
| logical channel subsystems for 3907, 3906, 2965, 2964, 2828, 2827, 2818, 2817, 2098, 2097, 2096, 2094, 2086,
and 2084 CPCs. It is intended for system programmers and service representatives who are responsible
for defining, installing, and configuring the channel paths, control units, and I/O devices on CPCs that
require ICP IOCP.
Table 1. EC levels that ICP IOCP supports
Supported models EC level
TM TM ™
| IBM z14 (z14 ) / IBM LinuxONE Rockhopper II (Rockhopper II) - All EC levels
| 3907
IBM z14TM (z14TM) / IBM LinuxONE Emperor™ II (Emperor II) - 3906 All EC levels
® ®
IBM z13s (z13s ) / IBM LinuxONE Rockhopper (Rockhopper) - 2965 All EC levels
® ®
IBM z13 (z13 ) / IBM LinuxONE Emperor (Emperor) - 2964 All EC levels
®
zEnterprise BC12 (zBC12) - 2828 All EC levels
zEnterprise EC12 (zEC12) - 2827 All EC levels
zEnterprise 114 (z114) - 2818 All EC levels
zEnterprise 196 (z196) - 2817 All EC levels
® ™
System z10 Business Class (z10 BC) - 2098 All EC levels
System z10 Enterprise Class (z10 EC) - 2097 All EC levels
® ®
System z9 Business Class (z9 BC) - 2096 All EC levels
System z9 Enterprise Class (z9 EC) - 2094 All EC levels
zSeries 890 (z890) - 2086 All EC levels
zSeries 990 (z990) - 2084 All EC levels
© Copyright IBM Corp. 2017, 2018 xiii

Table 1. EC levels that ICP IOCP supports (continued)
Supported models EC level
Legend:
| z14 / Rockhopper II: Models ZR1 and LR1
z14 / Emperor II: Models M01, M02, M03, M04, M05, LM1, LM2, LM3, LM4, and LM5
z13s / Rockhopper: Models N10, N20, L10, and L20
z13 / Emperor: Models N30, N63, N96, NC9, NE1, L30, L63, L96, LC9, and LE1
zEnterprise BC12 - 2828: Models H06 and H13
zEnterprise EC12 - 2827: Models H20, H43, H66, H89, and HA1
zEnterprise 196 - 2817: Models M15, M32, M49, M66, and M80
zEnterprise 114 - 2818: Models M05 and M10
z10 BC - 2098: Model E10
z10 EC - 2097: Models E12, E26, E40, E56, and E64
z9 EC - 2094: Models S08, S18, S28, S38, and S54
z9 BC - 2096: Models R07 and S07
z990 - 2084: Models A08, B16, C24, and D32
z890 - 2086: Model A04
This user's guide assumes that the reader has a knowledge of the CPCs that support the characteristics of
their I/O resources.
Use IOCP to define I/O configuration data the CPC requires to control I/O requests. This user's guide
does not include information about how to configure I/O resources, but only on how to define the
configuration by using IOCP.
Note: For information describing the ESCON and FICON channel-to-channel (CTC) implementation, see
the ESCON and FICON Channel-to-Channel Reference, SB10-7034.
What is included in this user's guide

This user's guide contains the following:
v Chapter 1, “IOCP introduction,” on page 1 presents an overview of IOCP and the I/O configuration
process.
v Chapter 2, “IOCP statements and functions,” on page 53 describes IOCP statements, how to code them,
and IOCP functions.
v Chapter 3, “Running the z/OS version of IOCP,” on page 111 explains how to run IOCP under control
of the z/OS® system control program.
v Chapter 4, “Running the VM version of IOCP,” on page 119 explains how to run IOCP under a VM
system control program.
v Chapter 5, “Running the z/VSE version of IOCP,” on page 131 explains how to run IOCP under
control of the VSE system control program.
v Chapter 6, “Running the stand-alone version of IOCP,” on page 139 describes how to run IOCP from
the Hardware Management Console.
xiv IOCP User's Guide

v Chapter 7, “IOCP configuration reports,” on page 151 shows the configuration reports that IOCP
produces.
v Chapter 8, “IOCP messages,” on page 171 lists the messages (with the prefix ICP and DMSICP) that
IOCP sends.
The appendixes in this user's guide include:

v Appendix A, “Coding IOCP statements,” on page 237 describes the rules for coding IOCP statements
and the notation this document uses to show the syntax of the statements.
v Appendix B, “Lists of statement input,” on page 241 shows sample lists of IOCP input files.
v Appendix C, “Characteristics of the I/O interface timeout function,” on page 243 contains detailed
information about the timeout function.
v Appendix D, “List of I/O devices and control units,” on page 245 shows the IOCP keyword values to
specify in the IODEVICE and CNTLUNIT statements.
v Appendix E, “Configuration examples,” on page 253 shows several figures and their accompanying
IOCP input files that illustrate I/O configuration examples.
v Appendix F, “Machine limits and rules,” on page 285 lists the machine limits and rules for the different
CPCs on which IOCP can run.
v Appendix G, “Notices,” on page 299 contains electronic emission notices, legal notices, and trademarks.
Revisions
A technical change from the previous edition of this document is indicated by a vertical line (|) to the left
of the change.
New publications do not usually contain any revision notations. However, for the convenience of those
who need to know the changes from the previous book, Input/Output Configuration Program User's Guide
for ICP IOCP, SB10-7163, revision notations have been added to illustrate these changes.
Related information
The following documents contain information about topics related to IOCP.
OS/390
v OS/390 Hardware Configuration Definition Planning, GC28-1750, explains the concepts, actions, and
general planning considerations needed to define and maintain an I/O configuration using the
Hardware Configuration Definition (HCD).
v OS/390 Hardware Configuration Definition User's Guide, SC28-1848, explains, along with online
information, how to use HCD to define I/O configurations and eligible device tables to OS/390®.
v OS/390 Hardware Configuration Definition Scenarios, SC28-1850, contains about 20 scenarios of common
hardware configuration tasks.
v OS/390 Hardware Definition Configuration Messages, GC28-1849.
v OS/390 HCD Reference Summary, SX22-0043, provides a short overview of HCD tasks, navigation
options, and information about online help. It also describes the relations among the configuration
objects and lists the actions and action codes for each object.
v MVS/ESA HCD and Dynamic I/O Reconfiguration Primer, SG24-4037.
z/OS and z/OS.e

v z/OS Hardware Configuration Definition Planning, GA22-7525, explains the concepts, actions, and general
planning considerations needed to define and maintain an I/O configuration using the Hardware
Configuration Definition (HCD).
About this publication xv

v z/OS Hardware Configuration Definition User's Guide, SC33-7988, explains, along with online information,
how to use HCD to define I/O configurations and eligible device tables to z/OS.
v z/OS Hardware Configuration Definition Scenarios, SC33-7987, contains about 20 scenarios of common
hardware configuration tasks.
v z/OS Hardware Configuration Definition Messages, SC33-7986.
v z/OS and z/VM Hardware Configuration Manager User's Guide, SC33-7989.
v z/OS HCD Reference Summary, SX33-9032, provides a short overview of HCD tasks, navigation options,
and information about online help. It also describes the relations among the configuration objects and
lists the actions and action codes for each object.
z/VM
v z/VM General Information , GC24-5944, describes the features of z/VM®.
v z/VM Planning and Administration, SC24-5948, describes how to generate a z/VM system on CPC and
perform dynamic I/O configuration changes.
v z/VM CP Command and Utility Reference, SC24-5967, describes the IOCP CMS utility.
v z/VM Dynamic I/O Configuration Planning and Administration, SC24-6044, describes the use of the
Hardware Configuration Definition (HCD).
v z/OS and z/VM Hardware Configuration Manager User's Guide, SC33-7989.
z/VSE
v z/VSE Planning, SC33-8301.
v z/VSE System Control Statements, SC33-8305.
3907 models
| v Hardware Management Console (HMC) and Support Element (SE) information can be found on the
| console help system, or on the IBM® Knowledge Center at http://www.ibm.com/support/
| knowledgecenter/ (Select Products on the navigation bar, and then select your server).
| v IBM Z PR/SM Planning Guide, SB10-7169
| v IBM Z Stand-Alone Input/Output Configuration Program User's Guide, SB10-7173
| v 3907 Installation Manual for Physical Planning (IMPP), GC28-6974.
3906 models
v Hardware Management Console (HMC) and Support Element (SE) information can be found on the
console help system, or on the IBM Knowledge Center at http://www.ibm.com/support/
knowledgecenter/ (Select Products on the navigation bar, and then select your server).
v IBM Z PR/SM Planning Guide, SB10-7169
v IBM Z Stand-Alone Input/Output Configuration Program User's Guide, SB10-7173
v 3906 Installation Manual for Physical Planning (IMPP), GC28-6965.
2965 models
v IBM z Systems PR/SM Planning Guide, SB10-7162
v IBM z Systems Stand-Alone Input/Output Configuration Program User's Guide, SB10-7166
v z13s Installation Manual for Physical Planning (IMPP), GC28-6938.
xvi IOCP User's Guide

2964 models
v IBM z Systems PR/SM Planning Guide, SB10-7162
v IBM z Systems Stand-Alone Input/Output Configuration Program User's Guide, SB10-7166
v z13 Installation Manual for Physical Planning (IMPP), GC28-6938.
2828 models
v The content from the following publications is now incorporated into the Hardware Management
Console (HMC) and Support Element (SE) (Version 2.12.1) help system:
– System z Hardware Management Console Operations Guide
– zEnterprise System Support Element Operations Guide
Hardware Management Console (HMC) and Support Element (SE) information can be found on the
v zEnterprise BC12 System Overview, SA22-1089
v zEnterprise System PR/SM Planning Guide, SB10-7156
v System z Stand-Alone Input/Output Configuration Program User's Guide, SB10-7152
v zEnterprise BC12 Installation Manual for Physical Planning (IMPP), GC28-6914.
2827 models
v zEnterprise EC12 System Overview, SA22-1088
v The content from the following publications is now incorporated into the Hardware Management
Console (HMC) and Support Element (SE) (Version 2.12.1) help system:
– System z Hardware Management Console Operations Guide
– zEnterprise System Support Element Operations Guide
v System z Hardware Management Console Operations Guide (Version 2.12.0), SC28-6919
v zEnterprise System Support Element Operations Guide (Version 2.12.0), SC28-6920
v zEnterprise EC12 Installation Manual for Physical Planning (IMPP), GC28-6914.
2818 models
v zEnterprise 114 System Overview, SA22-1087
v zEnterprise 114 Installation Manual for Physical Planning (IMPP), GC28-6907.
2817 models
v zEnterprise EC12 System Overview, SA22-1086
About this publication xvii

v System z10 and System z9 Stand-Alone Input/Output Configuration Program User's Guide, SB10-7152
v zEnterprise EC12 Installation Manual for Physical Planning (IMPP), GC28-6897.
2098 models
v System z10 Business Class System Overview, SA22-1085
v System z10 Business Class Processor Resource/Systems Manager Planning Guide, SB10-7153
v System z10 and System z9 Stand-Alone Input/Output Configuration Program User’s Guide, SB10-7152
v System z Hardware Management Console Operations Guide, SC28-6873
v System z10 Business Class Support Element Operations Guide, SC28-6879
v System z10 Business Class Installation Manual for Physical Planning, GC28-6875.
2097 models
v System z10 Enterprise Class System Overview, SA22-1084
v System z10 Enterprise Class Processor Resource/Systems Manager Planning Guide, SB10-7153
v System z10 and System z9 Stand-Alone Input/Output Configuration Program User’s Guide, SB10-7152
v System z10 Enterprise Class Support Element Operations Guide, SC28-6868
v System z10 Enterprise Class Installation Manual for Physical Planning, GC28-6865.
2096 models
v System z9 Business Class System Overview, SA22-1083
v System z9 Processor Resource Systems Manager Planning Guide, SB10-7153
v System z9 Stand-Alone Input/Output Configuration Program User’s Guide, SB10-7152
v System z9 Support Element Operations Guide, SC28-6868
v System z9 Business Class Installation Manual for Physical Planning, GC28-6855.
2094 models
v System z9 Enterprise Class System Overview, SA22-6833.
v System z9 Processor Resource Systems Manager Planning Guide, SB10-7041
v System z9 Stand-Alone Input/Output Configuration Program User’s Guide , SB10-7152
v System z9 Support Element Operations Guide, SC28-6858
v System z9 Enterprise Class Installation Manual for Physical Planning, GC28-6844.
2086 models
v zSeries 890 System Overview, SA22-6832.
v zSeries 890 and 990 Processor Resource Systems Manager Planning Guide, SB10-7036.
v zSeries 890 and 990 Stand-Alone IOCP User's Guide, SB10-7040.
v Hardware Management Console Operations Guide, SC28-6830.
v zSeries 890 and 990 Support Element Operations Guide, SC28-6831.
xviii IOCP User's Guide

2084 models
v zSeries 990 System Overview, SA22-1032.
v zSeries 890 and 990 Processor Resource Systems Manager Planning Guide, SB10-7036.
v zSeries 890 and 990 Stand-Alone IOCP User's Guide, SB10-7040.
v Hardware Management Console Operations Guide, SC28-6819.
v zSeries 990 Support Element Operations Guide, SC28-6820.
v zSeries 990 Installation Manual for Physical Planning, GC28-6824.
Enterprise systems connection (ESCON)

v Introducing Enterprise Systems Connection, GA23-0383.
v Introducing ESCON Directors, GA23-0363.
v Planning for Enterprise Systems Connection Directors, GA23-0364.
v ESCON and FICON Channel-to-Channel Reference, SB10-7034.
v Planning for the 9032 Model 5 Director with FICON Converter Feature, SA22-7415.
Fibre connection (FICON)

v Introduction to IBM S/390 FICON, SG24-5176.
v Planning for the 9032 Model 5 Director with FICON Converter Feature, SA22-7415.
v ESCON and FICON Channel-to-Channel Reference, SB10-7034.
Open Systems Adapter (OSA)

v Open Systems Adapter-Express Customer's Guide and Reference, SA22-7935.
v OS/390 Open Systems Adapter Support Facility Users Guide, SC28-1855.
v VSE/Enterprise Systems Architecture Open Systems Adapter Support Facility Users Guide, SC28-1946.
v Open Systems Adapter-Express Integrated Console Controller User's Guide, SA22-7990.
Note to all users

This user's guide describes ICP IOCP. See Table 16 on page 285 for a list of the Central Processor
Complex (CPC) models that require ICP IOCP. Any IOCDS that ICP IOCP creates runs only on the CPC
models listed there. You cannot use ICP IOCP to print reports or read IOCDSs produced by IYP IOCP,
IZP IOCP, or IXP IOCP.
IOCP requires a maximum of 214 MB of storage to run on z/OS and z/VSE® (MB equals 1 048 576
bytes). On VM, IOCP uses 214 MB of storage in a CMS virtual machine. IOCP requires a variable amount
of storage based on the type of IOCDS it is processing. To determine the amount of storage IOCP requires
for a specific type of CPC model, see Table 18 on page 287.
When you perform any IOCP function, the z/OS IOCP program must be an APF-authorized program.
Accessibility
Accessible publications for this product are offered in EPUB format and can be downloaded from
Resource Link® at http://www.ibm.com/servers/resourcelink.
If you experience any difficulty with the accessibility of any IBM Z® and IBM LinuxONE information, go
to Resource Link at http://www.ibm.com/servers/resourcelink and click Feedback from the navigation
bar on the left. In the Comments input area, state your question or comment, the publication title and
number, choose General comment as the category and click Submit. You can also send an email to
[email protected] providing the same information.
About this publication xix

When you send information to IBM, you grant IBM a nonexclusive right to use or distribute the
information in any way it believes appropriate without incurring any obligation to you.
Accessibility features
The following list includes the major accessibility features in IBM Z and IBM LinuxONE documentation,
and on the Hardware Management Console and Support Element console:
v Keyboard-only operation
v Interfaces that are commonly used by screen readers
v Customizable display attributes such as color, contrast, and font size
v Communication of information independent of color
v Interfaces commonly used by screen magnifiers
v Interfaces that are free of flashing lights that could induce seizures due to photo-sensitivity.
Keyboard navigation
This product uses standard Microsoft Windows navigation keys.
Consult assistive technologies

Assistive technology products such as screen readers function with our publications, the Hardware
Management Console, and the Support Element console. Consult the product information for the specific
assistive technology product that is used to access the EPUB format publication or console.
IBM and accessibility

See http://www.ibm.com/able for more information about the commitment that IBM has to accessibility.
Related HMC and SE console information

knowledgecenter/. Select Products, and then select your server.
How to send your comments

Your feedback is important in helping to provide the most accurate and high-quality information. Send
your comments by using Resource Link at http://www.ibm.com/servers/resourcelink. Click Feedback on the
Navigation bar on the left. You can also send an email to [email protected]. Be sure to include the
name of the book, the form number of the book, the version of the book, if applicable, and the specific
location of the text you are commenting on (for example, a page number, table number, or a heading).
xx IOCP User's Guide

Summary of changes
Summary of Changes for SB10-7172-01 IBM Z Input/Output Configuration Program User's Guide ICP
IOCP
This revision contains technical and editorial changes.
New Information:
v Support for z14 model ZR1 and IBM LinuxONE Rockhopper II model LR1. See “Machine limits” on
page 286.
Summary of Changes for SB10-7172-00a IBM Z Input/Output Configuration Program User's Guide ICP
IOCP
New Information:
v Support for IBM LinuxONE Emperor II models (LM1, LM2, LM3, LM4, and LM5). See “Machine
limits” on page 286.
Summary of Changes for SB10-7172-00 IBM Z Input/Output Configuration Program User's Guide ICP
IOCP
New Information:
v New ICP IOCP release level: ICP IOCP Version 5 Release 4.0
v Support for Coupling Express® LR channel paths (TYPE=CL5; See “CHPID” on page 63) and was
introduced with ICP IOCP Version 5 Release 3.0.
v Support for z14 models (See “Machine limits” on page 286).
v Support for IBM zHyperLink Express adapters (See “FUNCTION” on page 103).
v Support for RoCE Express2 networking Ethernet PCIe feature (See “FUNCTION” on page 103).
v Support for MIXTYPE keyword on CHPID statement.
v Introduce FID summary report (See “FID summary report” on page 155).
Summary of Changes for SB10-7163-02 z Systems® Input/Output Configuration Program User's Guide
ICP IOCP
New Information:
v Support for UUID statement
v Support for UID keyword on FUNCTION statement
Summary of Changes for SB10-7163-01 z Systems Input/Output Configuration Program User's Guide
ICP IOCP
© Copyright IBM Corp. 2017, 2018 xxi

New Information:
v Support for z13s / Rockhopper 2965 models (See “Machine limits” on page 286)
v Support for Emperor 2964 models (See “Machine limits” on page 286
v Support for SMC-D virtual network connections (See “FUNCTION” on page 103)
v Support for Regional Crypto Enablement PCIe feature (See “FUNCTION” on page 103)
Summary of Changes for SB10-7163-00 z Systems Input/Output Configuration Program User's Guide
ICP IOCP
New Information:
v Support for IBM z13 2964 models (See Appendix F, “Machine limits and rules,” on page 285)
– Increase to 6 CSS, 85 partitions, 4 Subchannel sets
– Increase to 256 Coupling channel paths
– Increase to 32k device maximum on FICON and FCP channel paths
– Increase in FUNCTION FID range.
v Support for new CHPID Type CS5 (See “CHPID” on page 63)
v Support for shared 10Gb RoCE Express networking Ethernet PCIe feature (See “FUNCTION” on page
103)
v Support for new VCHID keyword on CHPID statement for Hipersocket (TYPE=IQD) channels.
Summary of Changes for SB10-7037-11 System z® Input/Output Configuration Program User's Guide
ICP IOCP
New Information:
v Support for the zEnterprise BC12 2828 models (See Appendix F, “Machine limits and rules,” on page
285)
v Support for the new FUNCTION statement to define new PCIe card functionality and logical
partition(s) (See “FUNCTION” on page 103)
v Support for 10Gb RoCE Express networking Ethernet PCIe feature
v Support for zEDC Express compression PCIe feature
v The 2828 models require power-on reset with an IOCDS written for the 2828 models. After an upgrade,
you cannot power-on reset with an IOCDS that was written for the prior CPC unless the 2828 IOCDS
was written in preparation of an upgrade.
Summary of Changes for SB10-7037-10 System z

Input/Output Configuration Program User's Guide ICP IOCP
New Information:
v Support for the 2827 models (See Appendix F, “Machine limits and rules,” on page 285.)
xxii IOCP User's Guide

v The 2827 models require power-on reset with an IOCDS written for the 2827. After an upgrade, you
cannot power-on reset with an IOCDS that was written for the prior CPC unless the 2827 IOCDS was
written in preparation of an upgrade.
Changed Information:
v ESCON is not supported on 2827 (and subsequent) models and should update to FICON.

New Information:
v Support for defining a HiperSocket CHPID (TYPE=IQD) for connecting to the intraensemble data
network (IEDN).
v Support for defining HiperSocket CHPIDs (TYPE=IQD) to be capable of being bridged to an external
user network.
v IOCP now expects all devices defined for an IQD CHPID to be device type (UNIT=) of IQD.
v The Hardware Management Console diskette drive is not available on the zEnterprise System.

New Information:
v Support for the OSA for zBX data network channel paths (TYPE=OSX)
v Support for the OSA for zBX management network channel paths (TYPE=OSM)
v Support for a third subchannel set within a CSS
v Support for up to 32 HiperSockets™ CHPIDs
v Support for 128 coupling CHPIDs
v LPAR=YES in the PARM parameter of the EXEC statement for z/OS and z/VSE and the CMS option
LPAR are no longer supported. IOCP always generates an LPAR mode IOCDS.

Summary of changes xxiii

New Information:
v New ICP IOCP release level (stand-alone and batch versions): ICP IOCP Version 2 Release 2.0
v Support for the 2098 models. IOCP ignores the MAXDEV keyword on the RESOURCE statement and
reserves the maximum number of devices for these models. Also, IOCP reserves the maximum number
of CSSs and logical partitions for these models.
v LPAR=YES in the PARM parameter of the EXEC statement for z/OS and VSE and the CMS option
LPAR are no longer supported. IOCP always generates an LPAR mode IOCDS.

New Information:
v Support for the 2097 models. IOCP ignores the MAXDEV keyword on the RESOURCE statement and
reserves the maximum number of devices for these models. Also, IOCP reserves the maximum number
of CSSs and logical partitions for these models.
v Support for coupling over InfiniBand® channel paths (TYPE=CIB)
v I/O configurations with reconfigurable FCP channel paths can reach the maximum number of FCP
valid subchannels for 2097 models. Prior models did not reach the maximum. See “Fibre Channel
Protocol support” on page 37 for more information about how valid subchannels are calculated for
reconfigurable FCP channel paths.
v The OS keyword on the CHPID statement is no longer supported (use CHPARM keyword instead).
Summary of Changes for SB10-7037-05

System z9 Business Class and Enterprise Class and
zSeries 890 and 990
New Information:
v Support for the 2096 models
v The System z9 Enterprise Class (z9 EC) was formerly the System z9 109.

System z9 109 and zSeries 890 and 990
This major revision contains technical and editorial changes.
xxiv IOCP User's Guide

New Information:
v Support for the 2094 models
v Support for the OSA for network control program (NCP) channel paths (TYPE=OSN)
v Support for two subchannel sets within a CSS, using the new SCHSET keyword on the IODEVICE
statement
v IOCP uses a variable amount of storage based on the type of IOCDS it is processing (see Table 18 on
page 287).

zSeries Input/Output Configuration Program User's Guide ICP IOCP
This minor revision contains technical and editorial changes.
New Information:
v Increase the number of FCV, FC, FCP, OSC, OSD, and OSE channels available on 2084 and 2086 models
(see Table 18 on page 287).

New Information:
v Support for multiple control units on an OSD channel path so a single logical partition can have more
than 254 devices (84 TCP/IP stacks).
v Support for 1920 valid subchannels (640 TCP/IP stacks) on OSA-Express2 OSD channel paths when
priority specification is disabled. See “OSA configuration considerations” on page 33.

New Information:
v Support for the 2086 models (see Appendix F, “Machine limits and rules,” on page 285)
v Support for OSA-Express integrated console controller channel paths (TYPE=OSC)
v Reserved logical partitions can be specified which do not exist until a dynamic I/O configuration
change is made to define them
v A null device candidate list can be specified for a CSS which excludes the device from the CSS
v The following channel path types can now be spanned across multiple CSSs: FC, FCP, CFS, CFP, CBS,
CBP, OSC, OSD, and OSE.
Summary of Changes for SB10-7037-00a

This minor revision contains technical and editorial changes.
New Information:
v Hexadecimal value for Spanned ICP or IQD channel paths (See Table 26 on page 296).
Summary of changes xxv

This edition is the initial edition of this publication.
New Information:
v Support for the 2084 models (See Appendix F, “Machine limits and rules,” on page 285)
xxvi IOCP User's Guide

Chapter 1. IOCP introduction
This chapter presents an overview of the following topics:
v Hardware I/O configuration
v I/O configuration process
v I/O configuration data sets (IOCDSs)
v I/O configuration considerations
v Shared ESCON CTC considerations
v Coupling facility support
v Open Systems Adapter (OSA) feature support
v Internal Queued Direct Communication (HiperSockets) support
v Fibre Channel Protocol support
v Defining FICON switch configurations
v Managing logical paths for ESCON and FICON channels
v Managing the establishment of logical paths
v Channel path selection
v Channel unit types
v Versions of IOCP
Hardware I/O configuration

In the CPC, the input/output (I/O) subsystem, which controls channel operations, requires specific data
about the hardware I/O configuration.
To define the I/O configuration for the I/O subsystem, run the Input/Output Configuration Program
(IOCP). To run IOCP, you need to specify:
v Logical partitions
v Channel paths on the central processor complex (CPC) and their assignment to logical partitions
v Control units attached to the channel paths
v I/O devices assigned to the control units
v PCIe adapter functions on the CPC and their assignment to logical partitions.
Also, to meet changing I/O requirements, you can replace an existing I/O configuration with a new
configuration by running IOCP.
All control units and I/O devices that attach to the CPC must be defined to the channel subsystem by
using IOCP.
See Appendix D, “List of I/O devices and control units,” on page 245 for a partial list of control units
and I/O devices available.
Note: Your CPC might not support all the devices listed in Appendix D, “List of I/O devices and control
units,” on page 245.
| 3907, 3906, 2965, 2964, 2828, 2827, 2818, and 2817 models
Figure 1 on page 2 and Figure 2 on page 3 show the major components of the input/output (I/O)
| subsystem for 3907, 3906, 2965, 2964, 2828, 2827, 2818, and 2817 models and suggests ways to configure
© Copyright IBM Corp. 2017, 2018 1

I/O for maximum availability.
S01 S38
S02 S37
S03 S36
S04 S35
PCIe S05 S34 PCIe

RII
S06 S33
S07 S32
S08 S31
S09 S30
CU1
S11 S28
S12 S27
S13 S26
S14 S25
PCIe S15 S24 PCIe

RII
S16 S23
CU2 S17 S22
S18 S21
S19 S20
Figure 1. I/O channel path configuration for a representative PCIe I/O drawer in a 2817, 2818, 2827, 2828, 2964,
2965, and 3906 model
2 IOCP User's Guide

|
S02 S12
S03 S13
CU2
S04 S14
S05 S15
PCIe S06 S16 PCIe

RII
S07 S17
S08 S18
S09 S19
S10 S20
CU1
|
| Figure 2. I/O channel path configuration for a representative PCIe I/O drawer in a 3907 model
|
Legend
PCIe PCIe interface
RII Redundant I/O Interconnect bus
S Slot
Table 2, Table 3, and Table 4 on page 4 show which I/O slots of an I/O drawer for 3907, 3906, 2965, and
2964 models are in the same resource group. Resource groups are discussed in “Availability guidelines for
| 3907, 3906, 2965, 2964, 2828, 2827, 2818, and 2817 models” on page 6.
| Table 2. Resource group and corresponding slots for 3907 model
| Resource Group Slots in Resource Group
| 1 2,4,7,9
| 2 12,14,17,19
| 3 3,5,8,10
| 4 13,15,18,20
|
Table 3. Resource group and corresponding slots for 3906 model
Resource Group Slots in Resource Group
1 1-9
2 30-38
3 11-19
Chapter 1. IOCP introduction 3

Table 3. Resource group and corresponding slots for 3906 model (continued)
4 20-28
Table 4. Resource group and corresponding slots for 2964 and 2965 models
1 1-19
2 20-38
Figure 3 on page 5 and Figure 4 on page 6 shows the major components of the input/output (I/O)
subsystem for 2827 and 2817 models and suggests ways to configure I/O for maximum availability.
4 IOCP User's Guide

S01 S32
S03 S31
IFB H128
S06 S30
IFB H228
S08 H105 IFB S29
S02 H205 IFB S26
S04 S24
S07 S21
S09 IFB H123 S19

CU1
S10 IFB H223 S27
S12 S25
S15 S22
S17 H114 IFB S20
S11 H214 IFB
S13
CU2
S16
S18
Figure 3. I/O channel path configuration for a representative I/O cage in a 2827 and 2817 model
Legend
IFB InfiniBand interface
H1 Half-high card in top half of slot
H2 Half-high card in bottom half of slot
S Slot

CU2
S08 S10 S07 S11
IFB D109 D209 IFB
S02 S05 S03 S04
CU1
Figure 4. I/O channel path configuration for a representative I/O drawer in a 2965, 2964, 2828, 2817, 2818, and 2827
model
Legend
D1 Daughter card in left side of slot
D2 Daughter card in right side of slot
S Slot
| Availability guidelines for 3907, 3906, 2965, 2964, 2828, 2827, 2818, and 2817
models
When configuring devices with multiple paths to the same CPC, select any of the channel paths shown in
Figure 3 on page 5 or Figure 4 that:
v Are available on the CPC you are defining
v Are the correct type (ESCON, FICON, and so on) to meet the control unit, coupling facility, or network
attachment requirements
v Satisfy the rules regarding the mixing of channel types to a control unit.
6 IOCP User's Guide

However, for maximum availability of the device, OSA network, coupling facility, or PCIe function on a
| 3907, 3906, 2965, 2964, 2828, 2827, 2818, or 2817 model, consider the following guidelines.
v Choose channels plugged in different I/O domains (see CU 1). An I/O domain in an I/O cage or I/O
drawer contains four channel cards controlled by a single IFB-MP card. For example, the domain for
the IFB-MP card in H105 in the I/O cage contains slots 01, 03, 06, and 08 and in D109 in the I/O
drawer contains slots 02, 05, 08, and 10. An I/O domain in a PCIe I/O drawer contains eight channel
cards controlled by a single PCIe-IN card. For example, the domain for the PCIe-IN card in slot 05 in
the PCIe I/O drawer contains slots 1-4 and 5-9.
Note: This action is also recommended for optimum performance of your most heavily used I/O
devices.
When choosing the I/O domains to use, whether from different cages/drawers or the same
cage/drawer, consider using a combination of I/O domains from different books or CPC drawers.
When you must use InfiniBand or PCIe links from the same book or CPC drawer , try to use
InfiniBand or PCIe links from different Host Channel Adapter (HCA) or PCIe fanout cards on that
book. Refer to your PCHID Report to determine which InfiniBand and PCIe links belong to which
fanout cards on a book. If you have multiple paths to the device and multiple domains available that
have the correct channel type, spread the paths across as many books and CPC drawers and fanout
cards as possible.
Redundant I/O Interconnect is a function that allows one IFB-MP or PCIe-IN to back up another
IFB-MP or PCIe-IN for a failure or repair. The IFB-MPs in slot 05 of the I/O cage back up each other.
The same is true for IFB-MPs in slots 14, 23, and 28. In the I/O drawer, the IFB-MP cards in slot 09
back up each other. The PCIe-IN cards in slots 05 and 34 of the PCIe I/O drawer back up each other.
The same is true for the PCIe-IN cards in slots 15 and 24. Therefore, in the event an InfiniBand or PCIe
cable, fanout card, or book fails, the remaining IFB-MP or PCIe-IN card controls both domains. Certain
failures (for example, the IFB-MP or PCIe-IN card) might prevent the redundant takeover, which is a
reason to spread your paths over multiple domains.
When configuring Coupling over InfiniBand (CIB) channel paths for the same target CPC or coupling
facility, use InfiniBand links that originate from different books or CPC drawers and different HCA
fanout cards on those books/drawers. Doing so eliminates the HCA fanout card and the book as a
single point of failure where all connectivity would be lost.
When configuring Coupling Short Reach Gen5 channels (Integrated Coupling Adapter (ICA SR)
channel paths for the same target CPC or coupling facility, use ICA links that originate from different
CPC drawers and different ICA fanout cards on those drawers. This eliminates the ICA fanout card
and the drawer as a single point of failure where all connectivity would be lost.
When configuring Coupling Express LR (CE LR) channel paths for the same target CPC or coupling
facility, use CE LR links that originate from different I/O drawers and different PCIe-IN fanout cards
on those drawers. This eliminates the PCIe-IN fanout card and the drawer as a single point of failure
where all connectivity would be lost.
v If you define multiple paths from the same InfiniBand or PCIe link, distribute paths across different
channel cards (see CU 2). Also, if you define multiple coupling facility channels to the same coupling
facility or to the same ESA image, distribute paths across different coupling facility fanout cards, ISC
mother cards, or ISC daughter cards.
v The following channel path and function types are associated with resource groups and managed by
the CPC: CL5 channel paths and ROCE, ROC2, and ZEDC functions. These channel paths and function
| types are associated with the resource groups shown in Table 2 on page 3, Table 3 on page 3, and
Table 4 on page 4. The same resource groups apply to all I/O drawers for the CPC. A resource group
can become unavailable (for example, during an LMC patch install) and all of its channel paths and
functions will become unavailable. When configuring CL5 channel paths for the same target CPC or
coupling facility, use CE LR links that originate from different resource groups. When configuring these
function adapter types, be aware that those which are in the same resource group can all become
unavailable at the same time.

| Note: The examples shown in Figure 1 on page 2, Figure 2 on page 3, Figure 3 on page 5 and Figure 4
on page 6 merely try to illustrate availability guidelines and do not represent overall optimal
configurations.
| The 3907, 3906, 2965, 2964, 2828, 2827, 2818, and 2817 CPCs have eliminated the concept of FICON
channels or channel cards having affinity to specific SAPs. Now each SAP handles FICON work more on
an on-demand basis. That is, as FICON work for any channel arrives, the next available SAP handles that
request. It does not matter if it is an outbound request or an inbound interrupt, the next available SAP
handles the FICON work. There is no requirement to configure for “SAP affinity,” because there is no
affinity for FICON. For the other channel types (ESCON, and so on), the 2828, 2827, 2818, and 2817 CPCs
automatically balance installed channel cards across all available SAPs. The CPC tries to assign an equal
number of each channel card type to each available SAP. All channels on a given I/O card are always in
the same SAP, but it is not predictable which I/O cards are assigned to which SAPs. There are two
exceptions to the previous statements. First, Host Channel Adapters (HCAs) used for coupling are always
given affinity to a SAP on the local book. Second, if an OSA channel is defined as OSD, OSM, or OSX or
a FICON channel is defined as FCP, these channels use QDIO architecture and, therefore, do not actually
use any SAP resource during normal operations. For all channel types simply follow the preceding
recommendations for configuring for RAS, and the SAPs handle the workload appropriately.
2098 and 2097 models

Figure 5 on page 9 shows the major components of the input/output (I/O) subsystem for 2097 models
and suggests ways to configure I/O for maximum availability.
8 IOCP User's Guide

S01 S32
S03 S31
IFB H128
S06 S30
IFB H228
S08 H105 IFB S29
S02 H205 IFB S26
S04 S24
S07 S21
S09 IFB H123 S19

CU1
S10 IFB H223 S27
S12 S25
S15 S22
S17 H114 IFB S20
S11 H214 IFB
S13
CU2
S16
S18
Figure 5. I/O channel path configuration for a representative I/O cage in a 2097 model
Legend
S Slot

Figure 6 shows the major components of the input/output (I/O) subsystem for 2098 models and suggests
ways to configure I/O for maximum availability.
CU2
S08 S10 S07 S11
IFB D109 D209 IFB
S02 S05 S03 S04
CU1
Figure 6. I/O channel path configuration for a representative I/O drawer in a 2098 model
Legend
D1 Daughter card in left side of slot
D2 Daughter card in right side of slot
S Slot
Availability guidelines for 2098 and 2097 models

Figure 5 on page 9 for 2097 or Figure 6 for 2098 that:
10 IOCP User's Guide

However, for maximum availability of the device, OSA network, or coupling facility on a 2097 or 2098
model, consider the following guidelines.
v Choose channels plugged in different I/O domains (see CU 1). An I/O domain contains four channel
cards controlled by a single IFB-MP card. For example, the domain for the IFB-MP card in H105
contains slots 01, 03, 06, and 08 on a 2097 and in D109 contains slots 02, 05, 08, 10 on a 2098.
Note: This action is also recommended for optimum performance of your most heavily used I/O
devices.
When choosing the I/O domains to use, whether from different cages/drawers or the same
cage/drawer, consider using a combination of I/O domains from different books. When you must use
InfiniBand links from the same book, try to use InfiniBand links from different Host Channel Adapter
(HCA) fanout cards on that book. Refer to your PCHID Report to determine which InfiniBand links
belong to which HCA fanout cards on a book. If you have multiple paths to the device and multiple
domains available that have the correct channel type, spread the paths across as many books and
HCAs as possible.
Redundant I/O Interconnect is a function that allows one IFB-MP to back up another IFB-MP for a
failure or repair. The IFB-MPs in slot 05 back up each other on a 2097. (The same is true for IFB-MPs in
slots 14, 23, and 28.) On a 2098, the IFB-MP cards in slot 09 back up each other. Therefore, in the event
an InfiniBand cable, fanout card, or book fails, the remaining IFB-MP card controls both domains.
Certain failures (for example, the IFB-MP card) might prevent the redundant takeover, which is a
reason to spread your paths over multiple domains.
When configuring ICB-4 channel paths for the same target CPC or coupling facility, use a combination
of ICB links from different books and different MBA fanout cards on those books. Doing so allows for
continued connectivity if you have to run degraded after a book or MBA fanout card failure.
facility, use InfiniBand links that originate from different books and different HCA fanout cards on
those books. Doing so eliminates the HCA fanout card and the book as a single point of failure where
all connectivity would be lost.
v If you define multiple paths from the same InfiniBand link, distribute paths across different channel
cards (see CU 2). Also, if you define multiple coupling facility channels to the same coupling facility or
to the same ESA image, distribute paths across different coupling facility fanout cards, ISC Mother
cards, or ISC daughter cards.
Note: The examples shown in Figure 5 on page 9 and Figure 6 on page 10 merely try to illustrate
availability guidelines and do not represent overall optimal configurations.
2097 and 2098 CPCs have eliminated the concept of FICON channels or channel cards having affinity to
specific SAPs. Now each SAP handles FICON work more on an on-demand basis. That is, as FICON
work for any channel arrives, the next available SAP handles that request. It does not matter if it is an
outbound request or an inbound interrupt, the next available SAP handles the FICON work. There is no
requirement to configure for “SAP affinity,” because there is no affinity for FICON. For the other channel
types (ESCON, and so on), the 2097 and 2098 CPCs automatically balance installed channel cards across
all available SAPs. The CPC tries to assign an equal number of each channel card type to each available
SAP. All channels on a given I/O card are always in the same SAP, but it is not predictable which I/O
cards are assigned to which SAPs. There are two exceptions to the previous statements. First, ICBs and
Host Channel Adapters (HCAs) used for coupling are always given affinity to a SAP on the local book.
Second, if an OSA channel is defined as OSD or a FICON channel is defined as FCP, these channels use
QDIO architecture and, therefore, do not actually use any SAP resource during normal operations. For all
channel types simply follow the preceding recommendations for configuring for RAS, and the SAPs
handle the workload appropriately.

This figure shows the major components of the input/output (I/O) subsystem for 2094 and 2096 models
and suggests ways to configure I/O for maximum availability.
S01 S32
S03 S31
STI H128
S06 S30
STI H228
S08 H105 STI S29
S02 H205 STI S26
S04 S24
S07 S21
S09 STI H123 S19

CU1
S10 STI H223 S27
S12 S25
S15 S22
S17 H114 STI S20
S11 H214 STI
S13
CU2
S16
S18
Figure 7. I/O channel path configuration for a representative I/O cage in a 2094 or 2096 model
Legend
STI Self-timed interface

S Slot

Figure 7 on page 12 that:
model, consider the following guidelines.
v Choose channels plugged in different I/O domains (see CU 1). An I/O domain contains four channel
cards controlled by a single STI-MP card. For example, the domain for the STI-MP card in H105
contains slots 01, 03, 06, and 08.
Note: Choose channels plugged into different I/O domains for optimum performance of your most
heavily used I/O devices.
When choosing the I/O domains to use, whether from different cages or the same cage, consider using
a combination of I/O domains from different books. When you must use STIs from the same book, try
to use STIs from different MBA fanout cards on that book. Each fanout card plugged into a book
provides an MBA. Refer to your PCHID Report to determine which STI links belong to which MBA
fanout cards on a book. If you have multiple paths to the device and multiple domains available that
have the correct channel type, is also advisable to spread the paths across as many books and MBAs as
possible.
Redundant I/O Interconnect is a function that allows one STI to back up another STI after a failure.
The STIs in slot 05 back up each other. The same is true for those STIs in slots 14, 23, and 28.
Therefore, in the event of an STI cable, fanout card, or book failure, the surviving STI controls both
domains. Certain failures (for example, the STI-MP card) might prevent the redundant takeover, which
is why it is advisable to spread your paths over multiple domains.
When configuring ICB channel paths for the same target CPC or coupling facility, use a combination of
STIs from different books and different MBA fanout cards on those books. Doing so allows for
continued connectivity if you have to run degraded after a book or MBA fanout card failure.
When configuring ICB-3 channel paths for the same target CPC or coupling facility, use links that
originate from different STI-3 extender cards in the I/O cage. This eliminates the STI-3 card as a single
point of failure.
facility, use InfiniBand links that originate from different books and different HCA fanout cards on
those books. Doing so eliminates the HCA fanout card and the book as a single point of failure where
all connectivity would be lost.
v If you define multiple paths from the same STI, distribute paths across different channel cards (see CU
2). Also, if you define multiple coupling facility channels to the same coupling facility or to the same
ESA image, distribute paths across different coupling facility fanout cards, ISC Mother cards, or ISC
daughter cards.
Note: Each example in Figure 7 on page 12 is only an attempt to illustrate an availability guideline and
does not represent an overall optimal configuration.
2094 and 2096 CPCs automatically balance installed channel cards across all available SAPs. The CPC
tries to assign an equal number of each channel card type (ESCON, FICON, and so on) to each available
SAP. All channels on a given I/O card are always in the same SAP, but it is not predictable which I/O
cards are assigned to which SAPs. There are two exceptions to the previous statements. First, ICBs are

always given affinity to a SAP on the local book. Second, if an OSA channel is defined as OSD or a
FICON channel is defined as FCP, these channels use QDIO architecture and, therefore, really do not use
any SAP resource during normal operations. Configure for normal RAS as described in the preceding
paragraphs, and the machine balances the channels across your available SAPs.

Figure 8 on page 15 shows the major components of the input/output (I/O) subsystem for 2084 and 2086
models and suggests ways to configure I/O for maximum availability.

S01 S32
S03 S31
STI H128
S06 S30
S08 H105 STI S29
S02 H205 STI S26
S04 S24
S07 S21
S09 STI H123 S19

CU1
S10 STI H223 S27
S12 S25
S15 S22
S17 H114 STI S20
S11 H214 STI
S13
CU2
S16
S18
Figure 8. I/O channel path configuration for a representative I/O cage in a 2084 or 2086 model
Legend
STI Self-timed interface
S Slot

Figure 7 on page 12 that:
model, consider the following guidelines. 2086 models have a maximum of one I/O cage and one book.
v For systems with multiple I/O cages, distribute paths across the I/O cages (not shown). When
choosing which self-timed interfaces (STIs) to use in different cages, use the guidelines in the following
bullet.
v For systems with multiple STIs, distribute paths across the STIs (see CU 1).
Note: You are also recommended to do so for optimum performance of your most heavily used I/O
devices.
When choosing the STIs to use, whether from different cages or the same cage, consider using a
combination of STIs from different books. When you must use STIs from the same book, try to use STIs
from different MBAs on that book. Refer to Table 6 on page 19 or your Physical Channel Identifier
(PCHID) Report to determine which STI links belong to which MBAs on a book. If you have multiple
paths to the device and multiple domains available that have the correct channel type, spreading the
paths across as many books and MBAs as possible is also advisable.
Connecting your critical devices this way ensures access to those devices while running in a degraded
mode. In the unlikely event of an MBA or book failure, you might be able to take advantage of
running degraded until the repair.
When configuring ICB or coupling CHPIDs for the same target CPC or coupling facility, use a
combination of STIs from different books and different MBAs on those books. Doing so allows for
continued connectivity if you have to run degraded after a book or MBA failure.
When configuring ICB-2 or ICB-3 channel paths for the same target CPC or coupling facility, use links
that originate from different STI cards in the I/O cage. Doing so eliminates the STI card as a single
point of failure.
v If you define multiple paths from the same STI, distribute paths across different channel cards (see CU
2). Also, if you define multiple coupling facility channels to the same coupling facility or to the same
ESA image, distribute paths across different coupling facility fanout cards, ISC Mother cards, or ISC
daughter cards.
Note: Each example in Figure 7 on page 12 is only an attempt to illustrate an availability guideline and
does not represent an overall optimal configuration.
2084 CPCs automatically balance installed channel cards across all available system assist processors
(SAPs). The CPC tries to assign an equal number of each channel card type (ESCON, FICON, and so on)
in each available SAP. All channels on a given I/O card are always in the same SAP, but it is not
predictable which I/O cards are assigned to which SAPs. Configure for normal RAS as described in the
preceding paragraphs, and the machine balances the channels across your available SAPs. 2086 models
have a single SAP so all I/O is performed by that SAP.
The following table illustrates the relationship between the book jack numbers and the MBA that
supports those STIs.
Table 5. Relationship of book jack numbers to MBAs
Book jack numbers MBA
J00, J01, J02, J03 MBA-0

Table 5. Relationship of book jack numbers to MBAs (continued)
Book jack numbers MBA
J04, J05, J06, J07 MBA-1
J08 J09, J10, J11 MBA-2
Using a PCHID report to help plan your I/O configuration

When planning an I/O configuration, the PCHID Report from the Order Process Configurator (e.Config)
is a valuable aid.
Use the following information to relate this report to the availability guidelines presented previously:
| v “Availability guidelines for 3907, 3906, 2965, 2964, 2828, 2827, 2818, and 2817 models” on page 6
v “Availability guidelines for 2098 and 2097 models” on page 10.
v “Availability guidelines for 2096 and 2094 models” on page 13
v “Availability guidelines for 2086 and 2084 models” on page 16
v Although Figure 9 on page 18 does not show it, for 2084 and 2086 models, the Book/Jack/MBA
column shows the book number and MBA number by which the STI and channel card are serviced.
The PCHID Report for 2094 or 2096 models reports “Fanout” instead of “MBA,” indicating the fanout
slot where the MBA fanout card is plugged. Each fanout card contains an MBA.
v As shown in Figure 9 on page 18, the PCHID Report for 2828, 2827, 2818, 2817, 2098, and 2097 models
have a Source column that lists the book slot, fanout slot, and jack by which each installed feature is
serviced. The fanout slot is where the HCA or PCIe fanout card is plugged.
v The Cage column indicates the number of cages installed and the location of the I/O and CPC cages
containing I/O. (CPC cages are shown only if ICB-4s or HCAs for coupling are part of the
configuration.) The example shows three cages:
– A CPC cage in the top of the A-frame (A25B)
– An I/O cage in the bottom of the A-frame (A01B)
– An I/O cage in the bottom of the Z-frame (Z01B).
v The Slot column shows the slot into which the channel card is plugged.
v The F/C column shows the feature code for the I/O type (ESCON, FICON, and so on) installed in that
location.
v The PCHID/Ports or AID column shows the PCHID assignment and the physical port (jack number)
on the card in the slot with that assignment. For an HCA for a coupling fanout card, the Adapter ID
(AID) assigned to the HCA is reported in this column.
Note: The CHPID Mapping Tool is available from Resource Link. If you use this tool to help you assign
PCHIDs to CHPIDs, it provides you with a new report with your CHPID assignments and PCHID
values.
Following is an example of a PCHID Report.

Sample PCHID report
CHPIDSTART
17867327 PCHID REPORT Jun 26,2008
Machine: 2097-E26 NEW1
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Source Cage Slot F/C PCHID/Ports or AID Comment
06/D7 A25B D706 0163 AID=0C
15/D7 A25B D715 0163 AID=1C
06/DA A25B DA06 3393 01E/J01 01F/J02
15/DA A25B DA15 3393 03E/J01 03F/J02
15/D9/J01 A01B 01 3321 100/D1 101/D2 102/D3 103/D4
06/D9/J01 A01B 02 3367 110/J00J01 111/J02J03
06/D9/J01 A01B 04 3321 130/D1 131/D2 132/D3 133/D4
15/D9/J01 A01B 06 2323 140/J00 141/J01 142/J02 143/J03

144/J04 145/J05 146/J06 147/J07
148/J08 149/J09 14A/J10 14B/J11
14C/J12 14D/J13
06/D9/J01 A01B 07 3321 150/D1 151/D2 152/D3 153/D4
15/D9/J01 A01B 08 2323 160/J00 161/J01 162/J02 163/J03

164/J04 165/J05 166/J06 167/J07
168/J08 169/J09 16A/J10 16B/J11
16C/J12 16D/J13 16E/J14
06/D9/J01 A01B 09 2323 170/J00 171/J01 172/J02 173/J03

174/J04 175/J05 176/J06 177/J07
178/J08 179/J09 17A/J10 17B/J11
17C/J12 17D/J13
06/D8/J01 A01B 10 3321 180/D1 181/D2 182/D3 183/D4
15/D8/J01 A01B 11 3321 190/D1 191/D2 192/D3 193/D4
06/D8/J01 A01B 12 3321 1A0/D1 1A1/D2 1A2/D3 1A3/D4
15/D8/J01 A01B 16 2323 1D0/J00 1D1/J01 1D2/J02 1D3/J03

1D4/J04 1D5/J05 1D6/J06 1D7/J07
1D8/J08 1D9/J09 1DA/J10 1DB/J11
1DC/J12 1DD/J13
06/D8/J01 A01B 17 2323 1E0/J00 1E1/J01 1E2/J02 1E3/J03

1E4/J04 1E5/J05 1E6/J06 1E7/J07
1E8/J08 1E9/J09 1EA/J10 1EB/J11
1EC/J12 1ED/J13
15/D8/J01 A01B 18 2323 1F0/J00 1F1/J01 1F2/J02 1F3/J03

1F4/J04 1F5/J05 1F6/J06 1F7/J07
1F8/J08 1F9/J09 1FA/J10 1FB/J11
1FC/J12 1FD/J13
15/D9/J02 Z01B 01 3367 300/J00J01 301/J02J03
06/D9/J02 Z01B 02 3321 310/D1 311/D2 312/D3 313/D4
Figure 9. Sample PCHID report (Part 1 of 2)

15/D9/J02 Z01B 03 3321 320/D1 321/D2 322/D3 323/D4
06/D9/J02 Z01B 04 2323 330/J00 331/J01 332/J02 333/J03

334/J04 335/J05 336/J06 337/J07
338/J08 339/J09 33A/J10 33B/J11
33C/J12 33D/J13
15/D9/J02 Z01B 06 3321 340/D1 341/D2 342/D3 343/D4
06/D9/J02 Z01B 07 2323 350/J00 351/J01 352/J02 353/J03

354/J04 355/J05 356/J06 357/J07
358/J08 359/J09 35A/J10 35B/J11
35C/J12 35D/J13
15/D9/J02 Z01B 08 2323 360/J00 361/J01 362/J02 363/J03

364/J04 365/J05 366/J06 367/J07
368/J08 369/J09 36A/J10 36B/J11
36C/J12 36D/J13 36E/J14
06/D9/J02 Z01B 09 2323 370/J00 371/J01 372/J02 373/J03

374/J04 375/J05 376/J06 377/J07
378/J08 379/J09 37A/J10 37B/J11
37C/J12 37D/J13
06/D8/J02 Z01B 10 3321 380/D1 381/D2 382/D3 383/D4
15/D8/J02 Z01B 11 3321 390/D1 391/D2 392/D3 393/D4
15/D8/J02 Z01B 13 3321 3B0/D1 3B1/D2 3B2/D3 3B3/D4
06/D8/J02 Z01B 15 2323 3C0/J00 3C1/J01 3C2/J02 3C3/J03

3C4/J04 3C5/J05 3C6/J06 3C7/J07
3C8/J08 3C9/J09 3CA/J10 3CB/J11
3CC/J12 3CD/J13
06/D8/J02 Z01B 17 2323 3E0/J00 3E1/J01 3E2/J02 3E3/J03

3E4/J04 3E5/J05 3E6/J06 3E7/J07
3E8/J08 3E9/J09 3EA/J10 3EB/J11
3EC/J12 3ED/J13
15/D8/J02 Z01B 18 2323 3F0/J00 3F1/J01 3F2/J02 3F3/J03

3F4/J04 3F5/J05 3F6/J06 3F7/J07
3F8/J08 3F9/J09 3FA/J10 3FB/J11
3FC/J12 3FD/J13
Legend:
Source Book Slot/Fanout Slot/Jack
A25B Top of A frame
A01B Bottom of A frame
Z01B Bottom of Z frame
3367 OSA Express3 1000BASE T
3321 FICON Express4 10 KM LX
2323 ESCON Channel 16 Ports
3393 ICB 4 Link
0163 HCA2
Figure 10. Sample PCHID report (Part 2 of 2)
I/O configuration process

Specify I/O configuration data to IOCP by coding the data as card-image statements read from an input
device. Use the following input devices depending on the version of IOCP:
Table 6. Input device by IOCP version
IOCP version Input Device
z/OS version Card deck, tape, and disk

Table 6. Input device by IOCP version (continued)
IOCP version Input Device
VM version Card deck, tape, and disk
z/VSE version Card deck, tape, and disk
Stand-Alone version Diskette or memory stick and Support Element console. You can modify
cards by using a full-screen editor. (The diskette drive is not available
on the zEnterprise System.)
Although the input methods differ, IOCP performs the same process to build a configuration definition
and to write reports. IOCP then stores the definition in an I/O configuration data set (IOCDS) on the
hard disk of the Support Element. The IOCDS is used by the I/O subsystem.
Card-image input to IOCP

Code the following card-image statements and use them to define your I/O configuration data when you
run IOCP. When IOCP processes these statements, it also produces reports that allow you to check the
configuration.
ID An optional statement that specifies identification information printed in the heading of IOCP
configuration reports. It is the source of the customer name for the IOCDS. It is also the source of
the customer name that displays on the Input/Output Configuration window of the Support
Element and on the Power-on Reset window of the Support Element.
RESOURCE
A required statement that specifies the logical partitions and logical channel subsystems in the
configuration.
CHPID
An optional statement that specifies the channel paths installed on the CPC.
CNTLUNIT
An optional statement that specifies the control units attached to the channel paths.
IODEVICE
An optional statement that specifies the I/O devices assigned to the control units.
UUID An optional statement that specifies which logical partitions require UUID values.
FUNCTION
An optional statement that specifies the FUNCTION (PCIe) features installed on the CPC.
The IOCP statements are fully described under “Specifying IOCP statements” on page 53.
Console input to IOCP

When you run the stand-alone version of IOCP, use the full-screen editor to modify an existing IOCDS or
generate a new one. For a full description of the stand-alone version of IOCP, see Chapter 6, “Running
the stand-alone version of IOCP,” on page 139.
I/O configuration data sets

Use the I/O configuration data sets (IOCDSs) to store different I/O configuration definitions for your
CPC.
The CPC has four IOCDS slots. The IOCDS slots are: A0, A1, A2, and A3. The IOCDS slots exist on the
Support Element (SE) hard disk.

The configuration data in the IOCDS is accessed and transformed to a data format that the Licensed
Machine Code (LMC) in the I/O subsystem uses when you perform a power-on reset by selecting
Power-on Reset from the Support Element console or Hardware Management Console.
The I/O subsystem then controls I/O requests and uses the configuration data from the IOCDS that you
specify for the power-on reset.
IOCP can read from any IOCDS produced by the version of IOCP covered in this document. However,
IOCP can write only to an IOCDS that is not write-protected.
You can write-protect each IOCDS individually and remove write-protection when you want to alter an
IOCDS by using Input/Output Configuration window of the Support Element. Write-protection prevents
you from accidentally using IOCP to write over and destroy the configuration data in an IOCDS. Before
writing an IOCDS, you must remove the write-protection from that IOCDS.
When defining a new configuration, you can perform the following actions:
v Use the configuration in an existing IOCDS as your operating configuration
v Run IOCP to generate and write a new I/O configuration to another IOCDS
v Write-protect the new IOCDS
v Perform a power-on reset (POR) and use the new configuration to test the system.
IOCDSs and CPC modes

IOCP creates IOCDSs for LPAR mode only. For a complete description of LPAR mode, see the Processor
Resource/Systems Manager Planning Guide.
I/O configuration considerations

The IOCP generation process requires as input an I/O configuration input file containing IOCP
statements. You can write an IOCDS with the z/OS, VM, z/VSE, or stand-alone version of IOCP.
Defining multiple logical channel subsystems

IOCP has been dramatically enhanced to allow the definition of an I/O configuration containing multiple,
logical channel subsystems. A logical channel subsystem (CSS) is a complete definition of logical
partitions, channels, control units, and devices as you might have defined for any previous CPC. The
difference with this level of IOCP and the CPCs it supports is that multiple CSSs can be defined with
each CSS containing logical partitions, channels, control units, and devices. In addition, some CPCs now
support multiple subchannel sets, which allows a CSS to have more than twice as many devices.
IOCP has added the CSS parameter to many traditional keywords to identify the logical channel
subsystem to which the keyword parameters apply. IOCP has introduced a new keyword, SCHSET, to the
IODEVICE statement to identify the subchannel set in a CSS to which a device belongs. See Chapter 2,
“IOCP statements and functions,” on page 53 for complete syntax rules for using the CSS parameter and
specifying the SCHSET keyword. The following list summarizes the new keywords and parameters in
IOCP in support of multiple logical channel subsystems and multiple subchannel sets:
RESOURCE statement
Previously the RESOURCE statement defined the logical partitions (LPs) and associated partition
numbers (now called MIF image IDs) to be included in the configuration. With this level of IOCP,
you prefix those values with a CSS ID. The LPs and MIF image IDs are defined in the specified
CSS.
Note: What was called “partition number” in previous levels of IOCP is now called MIF image
ID. It retains all the same characteristics (for example, a hex number from 1-F), and is used the
same for CTC control unit definitions. However, for CPCs that ICP IOCP supports, the MIF

image ID is no longer used in the PARTITION keyword of the CF statement in the policy
information defined for the site's coupling facility resource management (CFRM) policy. Instead,
the “partition ID” that is specified on the activation profile for the CF image on the Support
Element identifies the CF image.
CHPID statement
The PATH parameter has traditionally been used to mirror the CHPID number assigned by the
CPC to the actual channel that is being defined. With ICP IOCP and the CPCs it supports, the
PATH parameter specifies a logical CHPID number of your choosing. The PATH parameter
allows the user to specify the CSSs and the CHPID number that is assigned to a channel path. If
it is a physical channel (for example, FICON), the new PCHID keyword identifies the location of
the physical channel in the CPC. For internal channels (for example internal coupling), the
PCHID keyword is not allowed, and only the logical CHPID number is required.
When a channel path is defined to multiple CSSs using the PATH parameter, the channel path is
defined as spanned. The PARTITION or NOTPART keywords can also use the CSS parameter to
indicate which LPs can access the channel in each CSS to which the channel path is assigned.
CNTLUNIT statement
The PATH and LINK keywords can have a CSS parameter that indicates the logical CHPIDs from
each CSS that have access to the control unit and the associated destination link address for each
path.
IODEVICE statement
The PATH keyword can have a CSS parameter to indicate a preferred path from each CSS for the
device. Likewise, the PARTITION or NOTPART keywords can have a CSS parameter to indicate a
unique device candidate list for each CSS that has access to the device. The SCHSET keyword
identifies the subchannel set in a CSS to which a device belongs; it can include a CSS parameter.
Note: If you are defining a single CSS configuration, the CSS parameter is optional.
Control unit headers

The I/O subsystem uses control unit headers to represent a set of physical control units that physically or
logically attach I/O devices in common. For example, a logical channel subsystem can use the control
unit header to establish a queue for I/O requests for the devices associated with the attaching physical
control units.
Note: Control unit headers were previously called logical control units.
IOCP builds control unit headers from the information specified in the CNTLUNIT and IODEVICE
statements. If a device attaches to more than one physical control unit, the physical control units form
part of (or the whole of) a control unit header.
IOCP builds a control unit header for the following control units:
v Each physical control unit with no I/O devices attached.
v Each physical control unit that does not share any I/O devices with other physical control units.
v Each group of two or more control units that share devices between them.
Figure 11 on page 23 shows examples of valid control unit headers.
When a physical control unit or a group of control units that share devices have channel paths in
multiple logical channel subsystems (CSSs), IOCP builds a control unit header (CUH) for each CSS. The
CUH in each CSS contains only the channel paths from the physical control units associated with that
CSS. The CUHs in each CSS are numbered independently of each other. The devices assigned to the
physical control unit or group of control units become subchannels in each CSS and are numbered
independently of each other. The subchannels in each CSS are associated with the CUH in the same CSS
and its channel paths. In Figure 11 on page 23, CUH 3 is in CSS 0 with paths 01 and 02. CUH 19 is in

CSS 1 with paths 21 and 24. A subchannel is generated in each CSS for device C. The subchannel in CSS
0 is associated with CUH 3 and the subchannel in CSS 1 is associated with CUH 19.
Control Unit Header 1 Control Unit Header 2
CHPID CHPID CHPID CHPID CHPID

01 01 07 02 04
Physical Physical Physical

Control Control Control
Unit 1 Unit 2 Unit 3
Device Device
A B
Control Unit Header 3 Control Unit Header 19
CSS 0 CSS 1
01 02 21 24 Control Unit Header 3 - CSS 0
with CHPIDS 01 and 02
Physical Physical Control Unit Header 19 - CSS 1

Control Control with CHPIDS 21 and 24
Unit 4 Unit 5
Device
C
Figure 11. Valid control unit headers
As it builds each control unit header, IOCP assigns a unique hexadecimal identifier to the control unit
header in each CSS. It also ensures that the control unit header is valid and that it complies with all the
following requirements:
v A maximum of eight physical control units are in one control unit header (CUH).
v A maximum of eight channel paths can attach to one CUH for each logical channel subsystem (CSS) in
a CPC.
v Each occurrence of a channel path counts as one channel path.
v Only one control unit can belong to a CUH if the control unit is an ESCON CTC or FICON CTC
control unit or if the control unit attaches to any of the following types of channel path:
– CBP
– CBS
– CFP

– CFS
– CIB
– CL5
– CS5
– EIO
– FCP
– ICP
– IQD
– OSC
– OSD
– OSE
– OSM
– OSN
– OSX
v Only one channel path per CSS can belong to a CUH if the control unit is an ESCON CTC or FICON
CTC or the control unit attaches to any of the following types of channel path:
– FCP
– IQD
– OSC
– OSD
– OSE
– OSM
– OSN
– OSX
If the control unit has channel paths in multiple CSSs, a CUH is built for each CSS with one path each.
v A physical control unit can belong to only one CUH in a CSS.
v Parallel physical control units within a CUH must all have the same characteristics (SHARED=Y or N;
PROTOCL=D, S, or S4).
v All channel path types (TYPE keyword) attaching to a CUH must meet the following restrictions:
– All CNC, FCV, or FC
– All CFS or CBS
– All CFP, CBP, ICP, CIB, CL5, or CS5
– All CBY or CVC
– All shared or all unshared (the SHARED, NOTPART, or PART or PARTITION keywords must be
consistently specified for all channel paths).
v A maximum number of control unit headers cannot be exceeded for each CSS, depending on the CPC
(see Appendix F, “Machine limits and rules,” on page 285).
Figure 12 on page 25 shows an example of a control unit header that is incorrect. Control unit header 5 is
incorrect because the physical control units have a combined total of more than eight channel paths in a
CSS.

Control Unit Header 5 for CSS 0
CHP CHP CHP CHP CHP CHP CHP CHP CHP

03 21 25 08 04 07 05 31 35
Physical Physical
Control Control
Unit 8 Unit 9
Device
5
Figure 12. Control unit header that is incorrect
Shared ESCON CTC considerations

IOCP can generate more unit addresses and unit address ranges for physical control units attaching to
shared ESCON CTC channel paths than for other channel path types or unshared channel paths. This
statement does not apply to FICON CTC control units.
Shared CTC unit address ranges

IOCP calculates the number of unit address ranges for a shared CTC channel path by performing the
following processing for each physical control unit attaching to the channel path:
v If a physical control unit has no devices, one unit address range is counted.
v If a physical control unit has devices, IOCP processes the candidate list for each device. (The candidate
list is determined by the PART, PARTITION, or NOTPART keyword in the IODEVICE statement.) A
device candidate list determines which logical partitions (LPs) can access the device. It can be
independent of the candidate list for the CTC channel path assigned to the device. IOCP determines
the number of unique LPs contained in the device candidate lists. The number of unique LPs is the
number of unit address ranges for the given physical control unit. The maximum number of unit
address ranges for a shared ESCON CTC channel path is 120.
If you do not specify the PART, PARTITION, or NOTPART keyword in an IODEVICE statement that is
assigned to a shared CTC channel path, the device candidate list defaults to the LPs in the candidate
list of the CTC channel path. Because the maximum number of physical control units and unit address
ranges for a shared CTC channel path are both 120, you cannot define 120 physical control units on a
shared CTC channel path if one of its control units has more than one unit address range.
To control the number of unit address ranges in a physical control unit assigned to a shared CTC
channel path, perform one or both of the following actions:

– Specify a CHPID candidate list that contains only the LPs that access the shared CTC channel path.
If the CHPID candidate list does not contain all the LPs in the configuration, devices that do not
specify a device candidate list have fewer unit address ranges for their physical control unit.
For example, if your configuration has 10 LPs and a CTC channel path is shared by 5 LPs in its
candidate list, each device assigned to the shared CTC channel path has a device candidate list that
defaults to 5 LPs. The physical control unit assigned to the shared CTC channel path has 5 unit
address ranges. Therefore, the maximum number of physical control units that you can define on
the shared CTC channel path is only 24 instead of 120 because this configuration first reaches the
maximum of 120 unit address ranges on the shared CTC channel path.
– Specify a device candidate list for each device assigned to a physical control unit. If the device
candidate lists of all the devices assigned to a physical control unit do not contain all the LPs in the
configuration, fewer unit address ranges are defined for the physical control unit.
For example, if your configuration has 10 LPs and all of them share a CTC channel path, each
device assigned to the shared CTC channel path has a device candidate list that defaults to all 10
LPs. If you specify device candidate lists so that only 4 unique LPs can access any of the devices on
a physical control unit, the control unit has 4 unit address ranges. Therefore, the maximum number
of physical control units that you can define on the shared CTC channel path is only 30 instead of
120 because this configuration first reaches the maximum of 120 unit address ranges on the shared
CTC channel path.
You can increase the number of physical control units you can define on a shared CTC channel path up
to the maximum of 120 by decreasing to one the number of unit address ranges on the control units
assigned to the CTC channel path.
Shared CTC unit addresses

IOCP calculates the number of unit addresses for a shared ESCON CTC channel path by multiplying the
number of unit addresses defined with the UNITADD keyword on the CNTLUNIT statement by the
number of unit address ranges for the physical control unit. (See “Shared CTC unit address ranges” on
page 25.) The maximum number of unit addresses for a shared ESCON CTC channel path is 512. Because
the maximum number of both unit addresses and devices for a shared CTC channel path is 512, you
cannot define 512 devices on a shared CTC channel path if one of its control units has more than one unit
address range.
For example, if your configuration has 8 LPs and a CTC channel path is shared by 4 LPs in its candidate
list, each device assigned to the shared CTC channel path has a device candidate list that defaults to 4
LPs. The physical control unit assigned to the shared CTC channel path has 4 unit address ranges. The
maximum number of unit addresses on the control unit is 128. Multiplying this number by the 4 unit
address ranges gives 512. Therefore, the maximum number of devices that you can define is only 128
because this configuration first reaches the maximum of 512 unit addresses on the shared CTC channel
path.
You can also control the number of devices per shared CTC channel path by specifying a device
candidate list for each device assigned to a control unit. However, the number of unit address ranges for
a control unit is based on the number of unique LPs in all the device candidate lists of the devices
assigned to the control unit.
You can increase the number of unit addresses and devices you can define on a shared CTC channel path
up to the maximum of 512 by decreasing to 1 the number of unit address ranges on each control unit
assigned to the CTC channel path.

Coupling facility support
IOCP supports the following coupling facility channel paths (TYPE keyword):
v CFS
v CFR
v CFP
v CBS
v CBR
v CBP
v ICP
v CIB
v CL5
v CS5
See Appendix F, “Machine limits and rules,” on page 285 to determine which channel path types your
CPC supports. You define coupling facility channel paths at the coupling facility and at the CPCs or
logical partitions (LPs) that connect to the coupling facility.
The coupling facility provides shared storage and shared storage management functions for the sysplex
(for example, high speed caching, list processing, and locking functions). Applications running on z/OS
images in the sysplex define the shared structures used in the coupling facility.
The coupling facility allows applications, running on multiple z/OS images that are configured in a
sysplex, to efficiently share data so that a transaction processing workload can be processed in parallel
across the sysplex.
Processor Resource/Systems Manager™ (PR/SM™) LPAR allows you to define the coupling facility, which
is a special logical partition (LP) that runs the coupling facility control code. The coupling facility control
code is Licensed Machine Code (LMC).
At LP activation, coupling facility control code automatically loads into the coupling facility LP from the
Support Element hard disk. No initial program load (IPL) of an operating system is necessary or
supported in the coupling facility LP.
Coupling facility control code runs in the coupling facility LP with minimal operator intervention.
Operator activity is confined to the CPC hardware operator console. PR/SM LPAR limits the hardware
operator controls typically available for LPs to avoid unnecessary operator activity.
Coupling facility channel hardware provides the connectivity required for data sharing between the
coupling facility and the CPCs directly attached to it. Coupling facility channels are point-to-point
connections that require a unique channel definition at each end of the channel.
Coupling facility channels can also connect two coupling facilities together for a duplexing configuration.
In this environment, coupling facility sender channels or coupling facility peer channels are configured to
coupling facilities to support communication between the coupling facilities. The following coupling
facility sections describe only ESA to coupling facility connections. However, they also apply to coupling
facility duplexing connections (CF to CF).
Coupling facility channels

Coupling facility channels are channels that provide the connectivity for data sharing between a coupling
facility and the CPCs or logical partitions (LPs) directly attached to it. IOCP provides controls for

defining coupling facility channels. However, IBM recommends that, when possible, you use the
Hardware Configuration Definition (HCD) to define the coupling facility channel configuration to the I/O
subsystem.
HCD provides controls for defining coupling facility channels. HCD also automatically generates the
control unit and device definitions associated with coupling facility sender and peer channel paths.
Coupling facility channels:

v Require a point-to-point connection (direct channel attach between a CPC or LP and a coupling
facility).
v Can connect a coupling facility to other LPs on the same CPC when a coupling facility is one of
multiple LPs running on a single CPC. Use internal coupling channels for these connections. Internal
Coupling channels can be used only to connect a coupling facility and LPs on the same CPC.
v Can be redundantly configured (two or more coupling facility channels from each CPC involved in
coupling facility data sharing) to enhance availability and avoid extended recovery time. This
statement does not apply to Internal Coupling channels.
v Require CFR or CBR channel path definition (coupling facility receiver channel capability) at the
coupling facility end of a coupling facility channel connection with a CFS or CBS channel path
definition (coupling facility sender channel capability) at the ESA end of a coupling facility channel
connection. You must use coupling facility sender and receiver channels when connecting to a CPC or
CF that does not support coupling facility peer channels (for example, a 9672).
v Require CFP, CBP, ICP, CIB, CL5, or CS5 channel path definition (coupling facility peer channel
capability) at the coupling facility and ESA ends of a coupling facility channel connection. Using
coupling facility peer channels is the recommended way to connect to a CPC or CF that does support
peer channels (for example, a 2084). Some CPCs that support peer channels do not support CIB
channels. Some CPCs that support CIB channels to not support CL5 and CS5 channels. CFP, CBP, ICP,
CL5, CS5, or CIB channel paths can be shared by many ESA LPs and only one coupling facility LP at
the same time.
v Require the following connections:
– The CFS channel path must be connected to a CFR channel path.
– The CBS channel path must be connected to a CBR channel path.
– The CFP channel path must be connected to another CFP channel path.
– The CBP channel path must be connected to another CBP channel path. ICB-3 and ICB-4 links are
both defined as CBP channel paths. Additionally, an ICB-3 link must be connected to another ICB-3
link, and an ICB-4 link must be connected to another ICB-4 link.
– The ICP channel path must be connected to another ICP channel path. Connect an ICP channel path
to another ICP channel path by specifying the CPATH keyword in the CHPID statement for every
ICP channel path.
– The CIB channel path must be connected to another CIB channel path. Connect a CIB channel path
to another CIB channel path by specifying the CSYSTEM and CPATH keywords in the CHPID
statement for every CIB channel path.
– The CS5 channel path must be connected to another CS5 channel path. Connect a CS5 channel path
to another CS5 channel path by specifying the CSYSTEM and CPATH keywords in the CHPID
statement for every CS5 channel path.
– The CL5 channel path must be connected to another CL5 channel path. Connect a CL5 channel path
to another CL5 channel path by specifying the CSYSTEM and CPATH keywords in the CHPID
statement for every CL5 channel path.
v Require a control unit definition and devices to communicate with the coupling facility end of a
coupling facility channel connection. Coupling facility peer channels connecting to only ESA ends of a
connection do not require a control unit nor devices. Each coupling facility sender channel requires two
devices. Each coupling facility peer channel defined to a control unit (and therefore connected to a CF
end of a connection) requires seven devices. When connecting two 2817 or later processors with Long

Reach HCA adapters (for example, HCA2-O LR), each CIB channel path in the control unit should
have 32 devices defined. When connecting Short Reach ICA adapters (for example, ICA SR), each CS5
channel path in the control unit should have 8 devices defined. When connecting Coupling Express LR
adapters, each CL5 channel path in the control unit should have a minimum of 32 devices defined but
IOCP tests for a minimum of eight.
You can configure coupling facility channels as follows:
Coupling facility sender channels (TYPE=CFS or TYPE=CBS)

You can configure a CFS and CBS channel path as:
v An unshared dedicated channel path to a single LP.
v An unshared reconfigurable channel path that can be configured to only one CF LP at a time but that
can be dynamically moved to another CF LP in the same CSS by channel path reconfiguration
commands
v A shared channel path that can be concurrently used by the LPs in the same CSS to which it is
configured.
v A spanned channel path that can be concurrently used by LPs in more than one CSS. Some CPCs do
not support spanned CFS and CBS channel paths. See “Keyword applicability for channel paths and
functions” on page 294.
CFS and CBS channel paths are used when connecting to a coupling facility that does not support
coupling facility peer channel paths. CFS channel paths are defined for ISC-3 links. CBS channel paths
are defined for ICB-2 links. Each CFS and CBS channel path should have two devices defined.
Coupling facility receiver channels (TYPE=CFR or TYPE=CBR)

You can configure a CFR and CBR channel path as:
v An unshared dedicated channel path to a single coupling facility LP.
v An unshared reconfigurable channel path that can be configured to only one coupling facility LP at a
time but that can be dynamically moved to another coupling facility LP in the same CSS by channel
path reconfiguration commands.
CFR and CBR channel paths are used when connecting to a CPC that does not support coupling facility
peer channel paths. CFR channel paths are defined for ISC-3 links. CBR channel paths are defined for
ICB-2 links.
Coupling facility peer channels (TYPE=CFP, TYPE=CBP, TYPE=ICP, TYPE=CIB,

TYPE=CL5, or TYPE=CS5)
You can configure a CFP, CBP, ICP, CIB, CL5, and CS5 channel path as:
v An unshared dedicated channel path to a single LP
v An unshared reconfigurable channel path that can be configured to only one LP at a time but that can
be dynamically moved to another LP in the same CSS by channel path reconfiguration commands
configured.
not support spanned CFP, CBP, and ICP channel paths. See “Keyword applicability for channel paths
and functions” on page 294.
Coupling facility peer channel paths can be shared by many ESA LPs and only one coupling facility LP at
the same time. CFP, CBP, CIB, CL5, and CS5 channel paths are used when connecting two CPCs that
support coupling facility peer channel paths. ICP channel paths are used when connecting coupling
facility and ESA LPs on the same CPC. CFP channel paths are defined for ISC-3 links. CBP channel paths
are defined for ICB-3 and ICB-4 links. CIB channel paths are defined for InfiniBand links. CS5 channel
paths are defined for Integrated Coupling Adapter links. CL5 channel paths are defined for Coupling

Express LR adapter links. Each CFP, CBP, ICP, CIB, CL5, and CS5 channel path assigned to a coupling
facility control unit should have seven, eight, or 32 devices defined.
Server Time Protocol: The Server Time Protocol (STP) provides time synchronization among CPCs. STP
uses coupling facility peer channels and provides a timing network between CPCs. STP can replace or
augment the Sysplex Timer. If your timing network is for a coupling sysplex, no special I/O definition is
required. STP operates over the existing coupling facility peer channel paths. However, if two systems
require a timing link but would not otherwise require coupling facility peer channel paths, a special I/O
definition is required. This definition consists of two coupling facility peer channels connected together
(one for each CPC). Each has a control unit definition with UNIT=STP specified and no devices.
Defining internal coupling channels (TYPE=ICP)

Internal Coupling (IC) channels are virtual attachments and, as such, require no real hardware. However,
they do require CHPID numbers and they do need to be defined in the IOCDS.
It is suggested that you define a minimum of Internal Coupling channels. For most customers, IBM
suggests defining just one pair of ICP channel paths for every two coupling facility logical partitions
(LPs) in your configuration. For instance, if your general-purpose configuration has several ESA LPs and
one CF LP, you would define one pair of connected ICP CHPIDs shared by all the LPs in your
configuration. If your configuration has several ESA LPs and two CF LPs, you still would define only one
connected pair of ICP CHPIDs, but define one ICP CHPID as shared by the ESA LPs and one CF LP and
define the other ICP CHPID as shared by the ESA LPs and the other CF LP. Both of these examples best
use the peer capabilities of these coupling channels by using the “sending” and “receiving” buffers of
both channels.
You must define an even number of ICP channel paths, and you must connect them in pairs. Use the
CPATH keyword in the CHPID statement to connect ICP CHPIDs. For more information about defining
Internal Coupling channels, see the Processor Resource/Systems Manager Planning Guide for your system.
Defining coupling over InfiniBand or PCIe-O (TYPE=CIB, TYPE=CL5, or TYPE CS5)

channels
InfiniBand (IFB) links on a Host Channel Adapter (HCA) used for coupling are defined as CHPID type
CIB. Ethernet links on a Coupling Express LR (CE LR) adapter used for coupling are defined as CHPID
type CL5. PCIe-O links on an Integrated Coupling Adapter (ICA) used for coupling are defined as
CHPID type CS5. Configurations with CIB, CL5, and CS5 CHPID definitions require the following unique
keywords:
LSYSTEM
is on the ID statement. The value of this keyword is the name you want to assign to the local CPC.
Use the same name as was specified for “CPC name” on the Hardware Management Console.
AID
is on the CHPID statement. This keyword is used only for CIB and CS5 channels. The value of this
keyword is the Adapter ID assigned to the Host Channel Adapter (HCA) containing the InfiniBand
link or the Adapter ID assigned to the Integrated Coupling Adapter (ICA) containing the PCIe-O link.
AID values are provided on the PCHID Report when an HCA or ICA is ordered (see “Using a
PCHID report to help plan your I/O configuration” on page 17) and on Hardware Management
Console panels. Instead of the AID keyword, CL5 channels use the PCHID keyword on the CHPID
statement as do most types of channel paths.
PORT
is on the CHPID statement. The value of this keyword is the port number for the InfiniBand link on
the HCA, the Ethernet link on the CE LR, or the PCIe-O link on the ICA.
CSYSTEM
is on the CHPID statement. This keyword is the name of the system with which this CHPID
connects. That is, it is the name that the LSYSTEM value in its IOCDS gives to the attached system.

CPATH
is on the CHPID statement. The value of this keyword, like ICP, identifies the CSS and CHPID on the
connecting system (CSYSTEM) to which this channel path is to communicate.
You can define up to 16 CHPIDs on an HCA, distributed any way between its ports. You can define up
to 4 CHPIDs per CE LR or ICA port for a total of 8 CHPIDs per CE LR or ICA. You might want to define
multiple CHPIDs on a physical link for the following reason: if one CPC has multiple coupling facility
(CF) images, you can define a CHPID for each CF image, reducing the requirement for physical links and
cables.
A configuration connecting two CIB CHPIDs or two CS5 CHPIDs within the same CPC using the same
AID and port is not supported. IOCP no longer allows this type of configuration. Also, a configuration
connecting two CL5 CHPIDs within the same CPC using the same PCHID and port is not supported.
Open Systems Adapter (OSA) feature support

The OSA features provide connectivity options that allow direct attachment of Fast Ethernet, Gigabit
Ethernet, 1000Base-T Ethernet, and Token Ring local area networks (LANs) to the CPC.
The OSA features, when defined as other than OSA-ICC and OSA for NCP, support the CPC acting as an
enterprise-wide server for multiple networks. LANs, LAN backbones, high speed workstations, hubs,
bridges, and routers that conform to these LAN interface standards can connect directly to the open
systems adapter, benefiting from the performance, connectivity, security, availability, and growth
opportunities associated with the z/Architecture®.
All OSA 1000Base-T Ethernet features, when defined as an OSA-Express Integrated Console Controller
(OSA-ICC), provide 3270 sessions that can serve as the system operator or master console for an
operating system so you can IPL. OSA-ICC replaces the 2074 Console Support Controller and local,
non-SNA 3174 control units. It uses TCP/IP over an Ethernet LAN to attach to personal computers which
are running TN3270E emulators. OSA-ICC also provides support for connecting to TPF through 3215 data
streams.
The OSA-Express2 or later Gigabit Ethernet and 1000Base-T Ethernet features, when defined as an OSA
for network control program (NCP) channel support traffic between an operating system image and a
Communication Controller for Linux (CCL) image. An OSA for NCP channel (TYPE=OSN) and CCL
replace the 374x control unit. An operating system image (for example, z/OS) communicates with the
OSN channel using the same Channel Data Link Control (CDLC) protocol it used when communicating
with the 374x control unit (for example, using an ESCON channel). The NCP function of the 374x control
unit is performed by CCL, which communicates with the OSN channel using the QDIO architecture. CCL
also has access to LANs through another OSA channel. This enables the operating system image to
communicate with the LANs connected to the CCL image. The operating system image communicates
with the OSN channel using devices defined in IOCP as 3745 (UNIT keyword on IODEVICE statement).
CCL communicates with the OSN channel using devices defined in IOCP as OSN.
The OSA-Express3 or later 10 Gigabit Ethernet features, when defined as an OSA for zBX data network
channel, supports the data traffic between the operating system and the zBX blade center. This defines
the connection to the zBX internal switch for the intraensemble data network (IEDN) providing physical
connection from zEnterprise to the blades.
The OSA-Express3 or later 1000BaseT Ethernet features, when defined as an OSA for zBX management
network channel, supports the management traffic between the operating system and the ensemble
management functions. This defines the connection to the intranode management network (INMN)
providing blade management functions.

OSA support facility
The OSA Support Facility for OS/390, z/OS, z/OS.e, z/VM, VSE/ESA, and z/VSE are applications that
provide configuration and operations management capabilities for the open systems adapters in the
configuration and allows you to install appropriate function onto open systems adapters to support client
server applications integrated at the CPC. The OSA Support Facility is not used for OSA-ICC or CHPID
types OSM and OSX. For OSA for NCP channels, the OSA Support Facility does not provide any
configuration management capabilities but provides capabilities only for operations management.
OSA channels
OSA channels allow:
v Direct connection of LANs using the Queued Direct I/O (QDIO) architecture
v Connection of LANs using non-QDIO architectures
v Attachment of TN3270E emulators for channels running the non-SNA DFT protocol (OSA-ICC)
v Connection to TPF through 3215 data streams (OSA-ICC)
v Communication between operating system images and a CCL image (OSA for NCP)
v Connection to the intraensemble data network (OSA for zBX data network)
v Connection to the intranode management network (OSA for zBX management)
All OSA features through the OSA-Express3 feature support two channel paths being defined in IOCP.
This is regardless of the number of ports on the feature, except for OSA-Express2 and OSA-Express3 10
Gigabit Ethernet, which supports only one channel path.
All OSA-Express4S features and later support a single channel path in IOCP regardless of the number of
ports on the feature.
Documents
The following documents provide additional information:
v Open Systems Adapter-Express Customer's Guide and Reference, SA22-7935.
v Open Systems Adapter-Express Integrated Console Controller User's Guide, SA22-7990.
OSA-ICC channels (TYPE=OSC)

All OSA features of the following type support OSA-ICC:
v 1000Base-T Ethernet.
OSA QDIO channels (TYPE=OSD)

The following OSA features support QDIO:
v OSA-Express
– Gigabit Ethernet
– 1000Base-T Ethernet
– Fast Ethernet
– Token Ring
v OSA-Express2 and OSA-Express3
– 10 Gigabit Ethernet
– 1000Base-T Ethernet.
v OSA-Express4S and later
– 10 Gigabit Ethernet

OSA non-QDIO channels (TYPE=OSE)
The following OSA features support non-QDIO:
v OSA-Express
– 1000Base-T Ethernet
– Fast Ethernet
– Token Ring
v OSA-Express2 and later
– 1000Base-T Ethernet.
OSA for zBX management network (TYPE=OSM)

Beginning with OSA-Express3, the following feature type supports the zBX management network:
v 1000Base-T Ethernet
OSA for NCP channels (TYPE=OSN)

The following OSA-Express2, OSA-Express3, OSA-Express4S, and OSA-Express5S features support NCP
emulation:
v Gigabit Ethernet
v 1000Base-T Ethernet.
OSA for zBX data network (TYPE=OSX)

Beginning with OSA-Express3, the following feature supports the zBX data network:
v 10 Gigabit Ethernet
OSA channels for OSA-ICC (TYPE=OSC), QDIO (TYPE=OSD), non-QDIO

(TYPE=OSE), zBX management network (TYPE=OSM), NCP
(TYPE=OSN), and zBX data network (TYPE=OSX)
You can configure an OSC, OSD, OSE, OSM, OSN, or OSX channel path as:
v An unshared dedicated channel path to a single logical partition (LP)
v An unshared reconfigurable channel path that can be configured to only one LP at a time but that can
be dynamically moved to another LP in the same CSS using channel path reconfiguration commands
configured
not support spanned OSC, OSD, and OSE channel paths. See “Keyword applicability for channel paths
and functions” on page 294.
OSA configuration considerations

IOCP supports the definition of open systems adapter (OSA) channels. OSA channels use either the
Queued Direct I/O (QDIO) architecture or non-QDIO architectures, or (based on how they are defined in
IOCP) they operate as one of the following:
v An OSA-ICC
v An OSA for NCP
v An OSA for a zBX data network
v An OSA for a zBX management network
The following table summarizes the connectivity the various channels provide and the channel path type
to use to define the channel types to the I/O subsystem.

Table 7. Summary of connectivity
Channel type Channel path type Connectivity the channel type provides
OSA QDIO OSD All LAN protocols supported by the OSA features that use
QDIO architecture
OSA non-QDIO OSE All LAN protocols supported by the OSA features that use
non-QDIO architectures
OSA-ICC OSC TN3270E emulators and 3215 data streams
OSA for NCP OSN CDLC protocol from an operating system image and QDIO
architecture from a CCL image. An OSA channel in the CCL
image provides LAN connectivity.
OSA zBX network OSM All LAN protocols supported by the OSA features that use
management QDIO architecture
OSA zBX data OSX All LAN protocols supported by the OSA features that use
management QDIO architecture
You define a control unit and devices for an OSC, OSD, OSE, OSM, OSN, or OSX channel path even
though no control unit or devices exist outside of the IOCDS definitions. (This is like ESCON and FICON
channel paths.) These device number and unit address definitions are associated with a channel path. For
OSD, OSE, OSM, and OSX channel paths, device numbers are used by software applications (for
example, VTAM®) to communicate with networks through an OSD, OSE, OSM, or OSX channel path. For
OSC channel paths, device numbers are necessary to configure a TN3270E emulator session or 3215 data
stream. For OSN channel paths, device numbers are necessary for an operating system image to
communicate with a CCL image. They also allow the operating system image to communicate with a
LAN connected to the CCL image.
For further information about OSD, OSE, and OSN configuration requirements for client server
applications, see Open Systems Adapter-Express Customer's Guide and Reference. For further information
about OSC configuration requirements, see Open Systems Adapter-Express Integrated Console Controller
User's Guide, SA22-7990.
Devices assigned to OSA channel paths are associated with a LAN port and each application using the
OSA channel card. A maximum number of OSA valid subchannels can be defined on OSD, OSM, OSN,
and OSX channels or used on OSE channels. An OSA valid subchannel is an OSD, OSE, OSM, OSN, or
OSX device with a unit address in the range 00-FD. The OSAD device with unit address FE (OSD, OSE,
OSM, OSN, or OSX only), used by the OSA Support Facility, is excluded from the maximum. Likewise, a
maximum number of valid subchannels can be used on OSC channels. A valid subchannel for an OSC
device can have any unit address in the range 00-FC.
Valid subchannels are devices that are defined to a logical partition and permitted for use by the logical
partition. Valid subchannels are calculated as follows:
v For unshared channel paths, each device defined in IOCP represents a single valid subchannel.
v For shared channel paths, each device defined in IOCP represents multiple valid subchannels based on
the total number of logical partitions (LPs) in its device candidate list.
For example, with a maximum of 480 valid subchannels for an OSD channel path with priority
specification enabled, if your configuration has 10 LPs and an OSD channel path is shared by all 10 LPs,
each device assigned to the shared OSD channel path has a device candidate list that defaults to all 10
LPs. This design allows you to define a maximum of 48 devices per LP for a total of 480 valid
subchannels on the channel path. However, if you specify device candidate lists of 5 LPs for each device
on the channel path, this design allows you to define a maximum of 96 devices per LP for a total of 480
valid subchannels.

TCP/IP assigns each OSA device on an OSD or OSX channel path four outbound queues, each of which
has a unique priority. These priorities allow the prioritization of outbound work based on TCP/IP Type
of Service (TOS) settings. If you choose not to use the priorities for a channel, the channel can support
four times as many devices where each device has a single outbound queue with a single priority. Specify
the CHPARM keyword for the OSD, OSM, or OSX channel path with bit 6 of the CHPARM value on
(CHPARM=02) to disable priority specification and support a maximum of 1920 OSA valid subchannels
(640 TCP/IP stacks). Otherwise, an OSD or OSX channel path has priority specification enabled and
supports a maximum of 480 OSA valid subchannels (160 TCP/IP stacks). OSM channel paths default to
disabled priority queuing. The only valid definition for OSM is CHPARM=02 (or omitting the CHPARM
keyword). Disabling priority specification is transparent to host policies and applications that set or affect
the setting of the TOS byte. For additional information about QDIO priority queuing and how it relates to
TOS and quality of service for Internet Protocol networks, see the z/OS Communications Server IP
Configuration Guide, SC31-8775.
Attention: Disabling priority specification (CHPARM=02) is supported only for OSA-Express2 and later
features. If it is disabled for an OSA-Express feature, the I/O subsystem flags the channel path with a
definition error and the channel path is unavailable until a dynamic I/O change is made or power-on
reset is performed with a new IOCDS to resolve the error.
Use of the OSA feature has the following considerations:

v For all OSA configurations:
– An OSA-Express, OSA-Express2 or later 10 Gigabit Ethernet channel path is associated with a single
port on an OSA channel card. An OSA-Express3 or later Gigabit and 1000Base-T Ethernet channel
path is associated with a pair of ports on the OSA channel card.
– For each OSA device on an OSD, OSE, OSM, or OSX channel path, specify UNIT=OSA on the
IODEVICE statement.
– Unit address FF is not allowed.
– The OSA feature using QDIO architecture uses the I/O device definitions and unit addresses in your
IOCDS.
– A maximum of 480 OSA valid subchannels (160 TCP/IP stacks) can be defined on an OSD or OSX
channel path when priority specification is enabled (CHPARM=00).
– A maximum of 1920 OSA valid subchannels (640 TCP/IP stacks) can be defined on an OSD channel
path on an OSA-Express2 or later feature, and on an OSM or OSX channel path on an OSA-Express3
or later feature when priority specification is disabled (CHPARM=02).
v For OSA configurations using the OSA support facility:
– Define a device with unit address FE for every OSD, OSE, or OSN channel path managed by the
OSA Support Facility, and specify UNIT=OSAD on the IODEVICE statement. For an OSD, OSM, or
OSX channel path, the control unit assigned to this device must either have a CUADD keyword
value of zero or have the CUADD keyword omitted. Otherwise, the device is unusable.
– A maximum of 240 OSA valid subchannels can be used on an OSE channel path even though IOCP
allows you to define more than this number. The OSA Support Facility manages this maximum.
v For OSE configurations not using the OSA support facility:
– An OSA channel card port configured as an OSE channel path has unit addresses 00-01 associated
with it when not using the OSA Support Facility.
– Only a single logical partition (LP) can use the OSE channel path. To ensure that the channel path
can only be used by a single LP, do one of the following actions:
- Define the channel path as dedicated or reconfigurable.
- Define the channel path as shared but give it access to only one LP or define both OSA devices
with the same, single LP in their device candidate lists.
- Define the channel path as shared and devices without device candidate lists but either configure
or activate host TCP/IP profiles for the OSA devices in only a single LP. At any time, only a
single LP can have a host TCP/IP profile active for the OSA devices.

v For OSC configurations:
– A maximum of 120 valid subchannels can be used on an OSC channel path even though IOCP
allows you to define more than this number. The Support Element manages this maximum.
– Unit address FE is not allowed.
– The OSC channel must be installed and defined as OSC and must be configured on the Support
Element using the Advanced Facilities window before TN3270E emulator sessions or 3215 data
streams can be used. Under Advanced Facilities, select either Panel Configuration Options or
Manual Configuration Options to configure each session you want to use.
– You can use an OSC channel path for either TN3270E emulator sessions or 3215 data streams but
not both. Only TPF supports 3215 data streams, and some CPCs require an RPQ be installed. Specify
the CHPARM keyword for the OSC channel path with bit 1 of the CHPARM value on
(CHPARM=40) to use 3215 data streams with TPF. Otherwise, the OSC channel path supports
TN3270E emulator sessions.
v For OSN configurations:
– For each OSA device on an OSN channel, you must specify UNIT=OSN or UNIT=3745 on the
IODEVICE statement.
– A maximum of 480 OSN valid subchannels and 180 3745 valid subchannels can be defined on an
OSN channel path.
– To reduce the number of valid subchannels, use the device candidate list to restrict the OSN devices
to CCL images and the 3745 devices to operating system images.
Internal queued direct communication (HiperSockets) support

HiperSockets provide a high speed internal TCP/IP connection between applications running on a CPC.
These applications (for example, z/OS Communication Server or Linux for zSeries) can be in different
z/VM images or different logical partitions. These “connections” are defined to the configuration with
Internal Queued Direct (IQD) communication channel path types.
A CPC can support 4096 separate data queues. Typically each queue is represented by one data device
and two control devices. IOCP ensures no more than 12288 (3*4096) valid subchannels are defined for all
the IQD channel paths in the configuration. Valid subchannels are devices that are defined to a logical
partition and permitted for use by the logical partition. Valid subchannels are calculated as follows:
v For unshared channel paths, each IQD device defined in IOCP represents a single valid subchannel.
v For shared channel paths, each IQD device defined in IOCP represents multiple valid subchannels
based on the total number of logical partitions (LPs) in its device candidate list.
All devices defined on an IQD channel path should be defined with a device type (UNIT=) of IQD.
A CPC allows up to 16 or 32 (see Appendix F, “Machine limits and rules,” on page 285 for your
processor) IQD channel paths. However, the only reason to define more than one is to ensure separation
of applications or if IQD special functions are required. If you require all applications in the CPC to
communicate with each other then only one IQD channel path is needed. If you require LPs in different
CSSs to communicate, then the IQD channel path must be spanned. If you require one set of LPs to
communicate with each other but not communicate with another set of LPs, then two channel paths need
to be defined with each LP set defined in one of the CHPID candidate lists.
You can define up to 12288 valid subchannels or devices on a single IQD channel path. Use as many
control unit (CNTLUNIT) definitions as needed to define the necessary number of devices.
You can define up to 256 devices per control unit. Each CNTLUNIT statement for the IQD channel path
must contain a unique CUADD value. A maximum of 64 control units can be defined per IQD channel
path.

The TCP/IP maximum transmission unit (MTU) and maximum frame size for internal queued direct
communication is specified using the CHPARM keyword on the CHPID statement. The following
CHPARM keyword values are supported:
v 00 for 8 KB MTU and 16 KB frame size
v C0 for 56 KB MTU and 64 KB frame size.
On 2828, 2827, 2817, and 2818 with enabling MCL applied or later processors, the following CHPARM
values are provided:
v 0x for 8 KB MTU and 16 KB frame size
v Cx for 56 KB MTU and 64 KB frame size
v x0 for a normal HiperSocket channel path
v x2 for a HiperSocket for IEDN channel path
v x4 for a HiperSocket capable to be externally bridged channel path.
Additional information about Internal Queued Direct Communication is in the z/OS Communications
Server IP Configuration Guide, SC31-8775.
Defining IQD channels

Internal queued direct (IQD) communication channels are virtual attachments and require no real
hardware. However, they do require CHPID numbers, and they do need to be defined in the IOCDS.
Also, beginning with the 2964 models, IQD channels require the VCHID keyword.
Define the minimum number of channel paths needed for separation of communicating LPs. Each IQD
channel path represents one internal LAN. If you have no requirement to separate LAN traffic between
your applications or are not enabling a special IQD function, only one IQD channel path needs to be
defined in the configuration.
Fibre Channel Protocol support

Fibre Channel Protocol (FCP) channels provide for the attachment of SCSI devices using the
industry-standard Fibre Channel Protocol for SCSI. Devices can be attached directly to the CPC using
point-to-point connections or through a storage area network (SAN). The SAN may consist of any
number of switches between the initiator (channel) and target (SCSI device) . Switch attachment for
targets may also include fibre channel arbitrated loops when supported by the switch. Devices with
parallel SCSI interfaces must be attached using fibre-channel-to-SCSI bridges.
A fibre channel is placed in the FCP mode of operation by defining the CHPID as type FCP in the I/O
configuration. Unlike FICON channels, the subchannels (IODEVICEs) on an FCP CHPID do not each
represent specific remote SCSI devices, but instead define an entry point to a SAN or a communications
link to a directly-attached SCSI device. The software uses the subchannel as the initiator of a connection
to a remote worldwide port name (WWPN) in order to address a specific SCSI logical unit number
(LUN). The remote WWPN and LUN are provided by the software.
An FCP channel path supports a maximum number of valid subchannels. Valid subchannels are devices
that are defined to a logical partition and permitted for use by the logical partition. Valid subchannels are
calculated as follows:
v For shared channel paths, each device defined in IOCP represents multiple valid subchannels based on
the total number of logical partitions (LPs) in its device candidate list.

For example, with a maximum of 480 valid subchannels, if your configuration has 10 LPs and an FCP
channel path is shared by all 10 LPs, each device assigned to the shared FCP channel path has a device
candidate to all LPs. This design allows you to define a maximum of 48 devices per LP for a total of 480
valid subchannels on the channel path. However, if you specify device candidate lists of 5 LPs for each
device on the channel path, this design allows you to define a maximum of 96 devices per LP for a total
of 480 valid subchannels.
Unit addresses FE and FF are reserved and cannot be assigned to an FCP subchannel.
FCP channels use the Queued Direct I/O (QDIO) interface and support the QDIO Assist functions of the
processor.
N_Port ID virtualization (NPIV)

By default, all subchannels on an FCP channel path use the same local WWPN to access the fibre channel
link. This means that SAN zoning and LUN masking cannot be used to manage the access rights of the
individual subchannels. In this mode, only one subchannel on the channel path may establish a
connection to a particular remote WWPN and LUN at a time. Permission to connect is granted on a
first-come, first-served basis to any subchannel on the channel path.
N_Port ID Virtualization (NPIV) is an industry-standard way to address this security issue. When NPIV
is used, each subchannel is assigned a unique local WWPN that it uses when the subchannel logs into the
SAN. The subchannel will be assigned a unique 3-byte fibre channel ID by the SAN and have all the
attributes of a separate physical fibre channel adapter. Consequently, each subchannel appears as a
different initiator to the SAN and target devices enabling the use of SAN zoning and LUN masking
access controls.
The NPIV mode of operation is not permitted in a point-to-point configuration.
When an FCP channel is shared among multiple partitions, the use of NPIV is configured separately for
each partition.
Because NPIV places additional resource requirements on both the SAN and the channel, IBM
recommends that the number of active subchannels on a single FCP channel be limited to a value
significantly lower than the configurable subchannel maximum. Care should be taken when defining the
SAN zone, as each NPIV subchannel will register for state-change notifications, significantly increasing
the number of messages the fabric must generate when a zone member goes through recovery, loses a
link, or logs in or out of the SAN. This and other recommended operational limits can be found in
Table 22 on page 293. Exceeding these limits may lead to inaccessible devices during error recovery
scenarios.
Defining FICON switch configurations

A fibre channel switch topology is called a fabric. A FICON fibre channel fabric can consist of one or more
switches. A FICON cascaded-switch fabric is a topology that consists of more than one switch between a
channel and its control unit.
When a FICON (TYPE=FC) channel connects to a switch, the SWITCH keyword must be specified on the
CHPID statement. The value specified is the switch number of the entry switch. The entry switch is the
switch to which the FICON channel actually connects. The switch number is an arbitrary value and is not
the same as the switch address. The switch number is required to be:
v In the range of 01-FF
v Unique from all other FICON switch numbers across the configuration
v Unique from any ESCON switch numbers specified for channel path types CVC, CBY, CNC, CTC, and
FCV.

Fibre channel port addressing identifies a destination port in a fibre channel fabric. Fibre channel port
addressing is always done with 3 bytes containing six hexadecimal digits, ddaapp. The first byte, dd, is the
switch domain ID or switch address. The switch address is set for a FICON switch by using switch
control panels. The second byte, aa, is the port area address or port address of the port on the switch.
This port (with its port address) is where a control unit or channel is connected and is known as an
F-Port. The third byte, pp, is an address on a fibre channel arbitrated loop. FICON does not support
arbitrated loops and for all FICON F-Ports in the fabric, the pp portion of the address must be a constant
value. The pp value is never specified in IOCP.
To support cascaded switches, the fabric must be what is called a high integrity fabric. A high integrity
fabric ensures data integrity by detecting any incorrect cabling within a FICON cascaded switch fabric
that could lead to I/O data being delivered to the wrong destination. A high integrity fabric is one that
both supports and is configured to use fabric-binding and insistent domain IDs. Fabric-binding is the ability
of the fabric to prevent another switch from being added operationally to the fabric without properly
planning for the new switch in the fabric and configuring the switch with the high integrity attributes.
An insistent domain ID is a function of the fabric that does not allow the switch address to be
automatically changed when a duplicate switch address is added to the fabric. Instead, fabrics using
insistent domain IDs require an operator to perform an overt action to change a switch address. The
customization of fabric-binding and the setting of an insistent domain ID are normally done only at
switch installation time or reinstallation time.
Note: Not all FICON switches support high integrity fabrics. Such switches can only be used in a
single-switch FICON fabric.
LINK address values for FICON control units are specified as 1 or 2 bytes as follows:
v When a channel path is connected to a cascaded-switch fabric and has a control unit not connected to
the same switch as the channel (that is, not the entry switch), 2-byte LINK address values must be
specified for the channel path in all of its control unit definitions. You do not need to change other
LINK address values in the control units. Control units can contain a mixture of 1-byte and 2-byte link
addresses.
v When all the control units for a channel path connect to the entry switch, the LINK address values for
the channel path in its control unit definitions can either be 1- or 2-byte addresses. Two-byte addresses
are supported only if the switch is configured with the high-integrity attributes.
If 2-byte link addresses are specified for a channel and it is not connected to a high-integrity fabric, the
channel fails to initialize and is unusable until one of the following actions occurs:
v The definitions are changed to 1-byte link addresses
v The fabric is configured to support the high-integrity attributes
v The channel connection is changed to a fabric that does support high-integrity.
A 2-byte LINK address value has a first byte with the switch address, dd, and a second byte with the port
address, aa. A one-byte LINK address value consists of only the port address, aa.
Rules for defining a cascaded FICON switch configuration follow:

v When a control unit is not connected to the entry switch for its channel path, the link address for that
path must be specified as a 2-byte address in the LINK keyword. Other paths to the control unit can
use 1- or 2-byte link addresses as appropriate.
v After a 2-byte address is specified for a channel path, all the link address values for that channel path
must also be specified as 2 bytes. This statement is true even if a control unit is attached to the entry
switch for the channel path.
The following are recommendations, but not requirements, when configuring switches for FICON:

v For an easy transition to a cascaded environment, after your switch fully supports and is customized
for high-integrity attributes, specify 2-byte link addresses for control units on that switch. This applies
even in a single switch configuration.
Note: Before converting to 2-byte link addresses, ensure that your applications and operating system
have any needed updates to support cascading.
v When configuring your switches and assigning switch addresses to them, use unique switch addresses
across all the fabrics in your entire enterprise and not only for the switches within a single fabric.
Note: A switch can support only certain switch addresses or domain IDs instead of the full range of
'00'-'FF'. For details, check the configuration information your switch vendor provides.
v When choosing the switch number (SWITCH keyword value) for your entry switch, try to use the
same value as the switch address of the switch to which the channel connects. Doing so reduces the
number of ways the same switch is known.
v After a 2-byte link address has been specified for a channel path, try to specify 2-byte link addresses
for all paths (from the same CPC) that have the same entry switch (SWITCH keyword). Doing so can
allow IOCP to perform better checking of the switch configuration.
Managing logical paths for ESCON and FICON channels

A logical path is a logical connection between a control unit and an ESCON channel (TYPE=CNC or CTC),
a FICON channel (TYPE=FC), or a FICON channel attached to an ESCON Director (TYPE=FCV). Logical
paths are important. For I/O operations to occur, a CPC requires the establishment of a logical path
between an ESCON or FICON channel and a control unit
Logical paths do not exist for ESCON channels attached to a 9034 ESCON Converter Model 1
(TYPE=CBY or CVC). These channels connect to a parallel channel port on a control unit. However, some
control units that support both parallel and ESCON channels reserve a logical path for each attached
parallel channel port.
Logical paths do not exist for:

v Coupling facility channel paths (TYPE=CFS, CFR, CFP, CBS, CBR, CBP, ICP, CIB, CL5, or CS5)
v OSA channels (TYPE=OSC, OSD, OSE, OSM, OSN, or OSX)
v Fiber channel protocol channels (TYPE=FCP),
v Internal queued direct communication channels (TYPE=IQD)
Control unit allocation of logical paths

Control units allocate logical paths to channels dynamically on a first-come-first-served basis. Control
units do not manage the allocation of logical paths. They allow channels to compete for logical paths
until all of the logical paths of a control unit are used.
Why manage logical paths?

The ESCON and FICON environments (the use of ESCON and FICON channels and ESCON and FICON
Directors) greatly enhances the connectivity potential for control units. In addition, you can define shared
channels that can request additional logical paths. However, control units can allocate only a limited
number of logical paths in relation to the number of logical paths that channels can request. (See Table 8.)
In configurations where channels request more logical paths than a control unit can allocate, you must
manage logical paths to ensure that the I/O operations you intend take place.
Table 8. Logical path summary by control unit
Control unit Maximum physical Logical paths per Maximum logical
links link paths
2105 storage controller

Table 8. Logical path summary by control unit (continued)
Control unit Maximum physical Logical paths per Maximum logical
links link paths
32 64 2048
2105
3990-2 16 Note 1 16
3990-3 16 Note 1 16
3990-6 16 Note 2 128
9343 D04 4 64 64
9343 DC4 4 64 64
3590 magnetic tape subsystem
3590 2 64 128
3490 magnetic tape subsystem
3490 D31, D32, D33, D34 1 16 16
3490 C10, C11, C12 1 16 16
3490 A01 2 16 32
3490 A10 4 16 64
3490 A02 4 16 64
3490 A20 8 16 128
3172 interconnect controller
3172-1 2 Note 3 16
3174 establishment controller
3174 12L (SNA only) 1 Note 3 8
3174 22L (SNA only) 1 Note 3 8
Notes:
1. 3990-2 and 3990-3 Storage Controls can have eight logical paths per side, two sides, with a minimum of one
logical path per adapter.
2. The 3990-6 Storage Control can have 64 logical paths per side, two sides, with a maximum of 32 logical paths
per adapter.
3. The 3172 Interconnect Controller and the 3174 Establishment Controller can have one logical path per logical
control unit (LCU) with a maximum of eight LCUs per link.
Managing the establishment of logical paths

You can manage the establishment of logical paths between channels and control units. With proper
planning, you can create I/O configuration definitions that allow control units in the configuration to
allocate logical paths for every possible request made by channels in either of the following ways:
v Create a one-to-one correspondence between the logical path capacity of all control units in the
physical configuration and the channels attempting to request them.
v Create I/O configurations that can exceed the logical path capacity of all or some of the control units
in the physical configuration but, at the same time, provide the capability to selectively establish logical
connectivity between control units and channels as needed.

Logical path considerations
You can better understand how to manage the establishment of logical paths by understanding the
following considerations:
v Control unit considerations
v Connectivity considerations
v Channel configuration considerations
v 9035 ESCON Converter considerations
Control unit considerations

Consider the following factors concerning the allocation of logical paths by control units:
v Control units allocate logical paths dynamically on a first-come-first-served basis.
v Control units do not manage the allocation of logical paths. They allow channels to compete for logical
paths until all the logical paths of the control unit are used.
v Control units vary in the number of logical paths they support. See Table 8 on page 40.
v Some control units that support a mix of parallel and ESCON ports can reduce the number of logical
paths allocated to ESCON channels because they reserve a logical path for each parallel port.
Connectivity considerations
ESCON and FICON system hardware, CPCs, and ESCON and FICON Directors, can significantly affect
the volume of logical path requests to a control unit as follows:
v Control units can attach to one or more ports on a director or to additional ports on other directors.
Each director port can dynamically connect to many other ports to which attached channels are
requesting logical paths.
v Each logical partition with channels attaching to the same control unit competes for the logical paths of
the same control unit.
v In a configuration where control units are shared by different CPCs, I/O configuration definitions for
individual control units are not coordinated automatically among the IOCDSs of the different CPCs.
Each CPC competes for the logical paths of a control unit.
v Shared channels require the establishment of a logical path for each active logical partition sharing the
channel. This requirement can significantly increase the number of logical paths that a single channel
requests.
Channel configuration considerations

The following configuration rules determine how logical paths are established for ESCON and FICON
channels.
v A channel initially attempts to establish logical paths:
– For channels configured online to logical partitions that are activated at POR. Shared channels
attempt to establish logical paths only for those activated logical partitions with the channel
configured online.
– When a logical partition is activated.
– When configured online (if previously configured offline).
v A channel that cannot initially establish a logical path can reattempt to establish a logical path if the
channel detects or is notified of:
– A change in the state of a control unit
– A change in the state of a link or port
– A dynamic I/O configuration change that frees previously allocated logical paths.
v A channel cannot establish a logical path or its logical path is removed if:
– A logical partition is deactivated. A shared channel continues to operate for any other remaining
activated logical partitions to which it is defined. Logical paths to those logical partitions remain
established.

– The director that connects the channel to the control unit blocks either the channel port or control
unit port used in the path.
– The director that connects the channel to the control unit prohibits the dynamic connection or
communication between the channel port and the control unit port used in the path.
– A link involved in the path fails or is disconnected. When a shared channel is affected by any of the
following circumstances, each logical partition sharing the channel is equally affected:
- The blocking of a port
- A prohibited dynamic connection or communication
- A failing or disconnected link
All logical paths using the port or link are removed (regardless of the logical partition with which
they are associated).
– The channel is configured offline. When a shared channel is configured offline for a logical partition,
it continues to operate for any other logical partition that has the channel configured online. Logical
paths to these other logical partitions remain established.
– Power to the channel, control units, or directors in the configuration is turned off.
v A shared channel cannot establish a logical path to a control unit for a logical partition that cannot
access any of the I/O devices on the control unit. The PART, PARTITION, or NOTPART keyword on
the IODEVICE statement specifies which logical partitions are permitted to access a device.
v A channel does not establish a logical path to a control unit if no devices are defined on the control
unit.
9035 ESCON converter considerations

For the 3990, 9035 configuration requires special Vital Product Data (VPD) specification. When an ESCON
Director blocks a port supporting a 9035 or prohibits a dynamic connection between the channel attached
to the 9035 and the control unit, the logical path is removed from the 3990. This path is now available to
other channels: either to another 9035 configured in the VPD or to channels not attached to a 9035 that
try to establish a logical path. The 3990 always tries to establish a logical path to the 9035s configured in
the VPD before it allows a channel not attached to a 9035 to have a logical path.
Recommendations for managing logical paths

You do not need to manage logical paths to a control unit if the control unit supports the number of
logical paths required by the I/O configuration. However, if a control unit does not support the number
of logical paths required by the I/O configuration, you can manage logical paths as follows:
v Deactivate unneeded logical partitions (LPs).
v Configure offline unneeded channels. For shared channels, configure offline unneeded channels on an
LP basis.
v For a control unit attached to shared channels, limit the number of LPs that can access the I/O devices
attached to the control unit.
To limit LP access to I/O devices, specify a device candidate list for each I/O device attached to the
control unit. The device candidate list (determined by the PART, PARTITION, or NOTPART keyword
in the IODEVICE statement) indicates the LPs that can access the I/O device. LPs not in the device
candidate list cannot access the I/O device.
A shared channel does not establish a logical path from an LP to an attached control unit if all the I/O
devices attached to the control unit deny access to the LP.
v Use the director to block ports or prohibit dynamic connections or communication between ports.
v Combinations of the preceding.

Channel path selection
A logical channel subsystem (CSS) uses a rotation order for the initiation of I/O requests to a device. In
addition, you can select a preferred path which causes the CSS to always attempt the initial selection on
that path.
Rotation order
You control the rotation order to a device with the order of the:
v Control unit numbers of the control units that attach to the device
v Channel paths used by the control units
Note: See “CNTLUNIT” on page 80 for a description of the CUNUMBR and PATH keywords in the
CNTLUNIT statement.
Rotation order does not apply to devices assigned to CFS, CFP, CBS, CBP, ICP, CIB, CL5, or CS5 channel
paths. The CSS selects the channel path that is likely to have the best response time.
The rotation order for a control unit header is built using the first channel path you specify for each
control unit that attaches to a device, beginning with the lowest numbered control unit. The rotation
order then is built using the second channel path, if any, for each control unit beginning with the lowest
numbered control unit. The rotation order is built by repeating this procedure until all channel paths for
all the control units contained in a control unit header have been added.
Control Unit Header 2 in Figure 11 on page 23 shows an example of how specifying paths in a control
unit header can affect the path rotation order. If you specify PATH=(01,07) in the CNTLUNIT statement
for physical control unit 2 and PATH=(02,04) for physical control unit 3, the rotation order for device B is
01, 02, 07, 04. However, if you specify PATH=(07,01) in the CNTLUNIT statement for control unit 2 and
PATH=(02,04) for control unit 3, the rotation order for device B is 07, 02, 01, 04.
Rotating path selection

The CSS tries to initiate each new I/O request for an I/O device on a rotating channel path basis within
the paths in the control unit header. The CSS tries the next channel path in the rotation order following
the last channel path initially selected for an I/O request for the control unit header. If that channel path
is busy or unavailable, the CSS chooses the channel path in the control unit header that has the greatest
likelihood of availability. Therefore, the rotation order is only used for the initial attempt for an I/O
request.
Path selection using a preferred path

When you specify a preferred path in the IODEVICE statement, the CSS always tries the preferred path
first. If it is busy or unavailable, the CSS chooses the channel path in the control unit header rotation
order that has the greatest likelihood of availability.
Channel paths to a device can belong to different logical partitions. I/O requests from any one logical
partition use only the paths you assign to that logical partition. The rotation order for a device in a
logical partition then follows the normal rotation order for the channel paths that belong to that logical
partition.
When you specify a preferred path in an IODEVICE statement that is available to more than one logical
partition, the device has a preferred path in all of its logical partitions in that CSS even though you can
specify only one channel path per CSS in an IODEVICE statement as a preferred path. The preferred path
for each LP is the first path in the rotation order that is defined to the LP beginning with the specified
preferred path.

For example, assume that Control Unit Header 2 in Figure 11 on page 23 has a rotation order of 01, 02,
07, 04 and paths 01 and 02 are in logical partition LP1 and paths 04 and 07 are in logical partition LP2. If
device B specifies 02 as its preferred path, then 02 is the preferred path for logical partition LP1, and 07 is
the preferred path for logical partition LP2.
Control unit types

IOCP defines control units as type 1 or type 2. Type 1 control units allow only a single channel program
at any one time. Type 2 control units allow multiple channel programs to run concurrently. IOCP uses the
SHARED parameter on the CNTLUNIT statement for control units assigned to only CVC channel paths
(TYPE keyword) to determine the control unit type as follows:
v For CVC (TYPE=CVC) channel paths:
– If you specify SHARED=Y, IOCP defines the control unit as type 1. The logical channel subsystem
allows only one I/O request at a time.
– If you specify SHARED=N, IOCP assigns the control unit as type 2. The logical channel subsystem
allows multiple I/O requests.
v For non-CVC channel paths:
– IOCP assigns all control units attached to CBY, CNC, CTC, EIO, FCV, FC, IQD, OSC, OSD, OSE,
OSM, OSN, or OSX channel paths as type 2. It ignores the SHARED keyword, if specified.
– IOCP assigns all control units attached to CBS, CBP, CFS, CFP, CIB, CL5, CS5 or ICP channel paths
as type 1. It ignores the SHARED keyword, if specified.
Note: Type 1 and type 2 control units as defined here pertain only to the number of channel programs
that can run concurrently. This statement is not directly related to the definitions of type 1, type 2, and
type 3 control units as given in System/360 and System/370 I/O Interface Channel to Control Unit Original
Equipment Manufacturers' Information, GA22-6974. However, specify as SHARED=Y all control units that
document GA22-6974 designates as type 1 or type 3. Control units that GA22-6974 designates as type 2
might be able to run one or multiple channel programs concurrently. Therefore, for IOCP purposes,
consider these to be type 1 or type 2.
Versions of IOCP
Four versions of IOCP are available; three are batch versions (or operating system versions), and one is a
stand-alone version (available on CPC Support Element). They are as follows:
v Batch versions (or Operating system versions)
– z/OS
– VM
– z/VSE
v Stand-Alone version
Batch versions of IOCP have an advantage over the stand-alone version. You can modify an IOCDS while
other system operations continue. Then, at a convenient time, perform the following steps:
1. Shut down your operating systems
2. Perform a power-on reset on the CPC (with the new IOCDS)
3. Activate your logical partitions
4. Reload your operating systems.
This method minimizes the amount of time that the system must be inactive while you reconfigure the
I/O devices.

Overview of the z/OS version of IOCP
The z/OS version of IOCP runs as a job under control of OS/390, z/OS, or z/OS.e. Run the z/OS
version of IOCP by using z/OS JCL statements. By coding options in the PARM parameter of the EXEC
statement, you can request the following IOCP functions:
v Generate an IOCDS.
IOCP reads, validates, and checks the syntax of the input statements. It generates an IOCDS based on
the configuration defined in the statements. It produces reports of the IOCDS generated, and, when
requested, IOCP writes the generated IOCDS to the Support Element hard disk.
v Produce configuration reports.
IOCP reads one or more IOCDSs into storage from the Support Element hard disk and prints the IOCP
configuration reports. (See Chapter 7, “IOCP configuration reports,” on page 151 for examples.)
See Chapter 3, “Running the z/OS version of IOCP,” on page 111 for a full description of the z/OS
version of IOCP. Figure 13 shows an overview of the processing the z/OS version of IOCP performs.
1 3
Input Statements IOCP Reports
IOCP
IOCDS in
Storage
4
IOCDSs
To HSA
Power-On Reset
1. The card-image input contains IOCP statements.
2. IOCP processes the IOCP statements, generates an IOCDS, produces reports, and (when requested) writes the
generated IOCDS to the Support Element hard disk.
3. The configuration reports allow you to check the I/O configuration as defined in the IOCDSs.
4. The IOCDSs reside on the Support Element hard disk. At power-on reset, the I/O configuration data is loaded
into the hardware system area (HSA). (Afterward, the I/O subsystem uses the data to control I/O requests.)
Figure 13. Overview of z/OS version of IOCP
IOCP and z/OS hardware configuration definition

The Hardware Configuration Definition (HCD) defines the I/O configuration information to both the
z/OS operating system and the CPC I/O subsystem. HCD also allows you to dynamically modify the
I/O configuration without performing:
v Power-on reset of the CPC
v Initial program load of the operating system.
Using HCD you can add, delete, or modify the I/O definitions of channel paths, control units, and I/O
devices in the CPC. It also allows you to save the changes you made to the I/O configuration definitions

and apply them to the power-on reset I/O configuration data set (POR IOCDS). You must use an IOCDS
generated from an I/O definition file (IODF) to be able to dynamically modify the I/O configuration. For
more information, see OS/390 Hardware Configuration Definition User's Guide, SC28-1848, or z/OS Hardware
Configuration Definition User's Guide, SC33-7988.
IOCDS write authorization

If you use HCD, you do not need to run IOCP directly to define the I/O configuration to the I/O
subsystem. However, HCD does run IOCP when it needs to write an IOCDS. You must be authorized to
write an IOCDS either through the system operator (IOCP message ICP050D prompts the operator) or
through a security product (for example, RACF®). To avoid operator intervention and the IOCP message
prompt (ICP050D), see “Authorizing an IOCDS write” on page 112.
Overview of the VM version of IOCP

The VM version of IOCP runs under the control of z/VM. Run the VM version of IOCP in the
conversational monitor system (CMS) environment. The CMS IOCP command runs IOCP and has options
that allow you to request the following IOCP functions:
IOCP reads, validates, and checks the syntax of the input statements. It generates an IOCDS based on
the configuration defined in the statements. It produces reports of the IOCDS generated, and, when
requested, IOCP writes the generated IOCDS to the Support Element hard disk.
IOCP reads one or more IOCDSs from the Support Element into storage and prints the IOCP
For a full description of the VM version of IOCP, see Chapter 4, “Running the VM version of IOCP,” on
page 119.
IOCP and system generation of a VM operating system

You must define your I/O configuration to the hardware and conditionally to VM. When you perform an
initial program load (IPL) of VM, VM creates its software I/O configuration by sensing the devices in the
hardware I/O configuration. If all the devices return all the information VM needs, VM completes its
software I/O configuration. If not, you must specify this information in the system configuration file or
the real I/O configuration file.
To define the devices in your I/O configuration that VM cannot sense, use a set of I/O statements for
your VM operating system. Use the system configuration file (SYSTEM CONFIG) on the parm disk. You
can also use the obsolete real I/O configuration file (HCPRIO) in the CP nucleus.
When you perform system generation, the control program uses the statements in the real I/O
configuration file to define the I/O configuration to the operating system. When you perform an IPL of
VM, VM uses a combination of the real I/O configuration file, the system configuration file, and the
devices it senses during IPL to define the I/O configuration to the operating system. VM also updates its
software I/O configuration by sensing devices when they are varied online or initialized.
To define your I/O configuration to hardware, you use a set of I/O statements in an IOCP file. When
you enter the IOCP command, the IOCP program uses the I/O statements in the IOCP file to define the
I/O configuration to the CPC.
For more information about performing these tasks, see Chapter 3, “Running the z/OS version of IOCP,”
on page 111. Figure 14 on page 48 shows an overview of the processing performed by the VM version of
IOCP.

1 4
2
SYSTEM Input Statements IOCP Reports

CONFIG
3
IOCP
IOCDS in
Storage
VM I/O 5
Configuration
IOCDSs
To HSA
Power-On Reset
1. Use the system configuration file (SYSTEM CONFIG) to define to the system control program those devices in
the I/O configuration that VM cannot sense. For VM system generation, you can also use the obsolete real I/O
configuration file (HCPRIO).
2. The input file contains the IOCP statements that define the I/O configuration to IOCP.
3. IOCP processes the IOCP statements, generates an IOCDS, produces reports, and, when requested, writes the
5. The IOCDSs reside on the Support Element hard disk. At power-on reset, the Support Element loads the I/O
configuration data into the hardware system area (HSA). (Afterward, the I/O subsystem uses the data in the
hardware system area (HSA) to control I/O requests.)
Figure 14. Overview of VM version of IOCP
IOCP and VM dynamic I/O configuration

VM allows you to dynamically modify the I/O configuration without performing:
v Power-on reset (POR) of the CPC
v Initial program load of the operating system.
On z/VM you can use HCD and HCM to add, delete, or modify the I/O definitions of channel paths,
control units, and I/O devices in the CPC. It also allows you to save the changes you made to the I/O
configuration definition data set (POR IOCDS). You must use an IOCDS generated from an I/O definition
file (IODF) to be able to dynamically modify the I/O configuration using HCD. For more information, see
z/VM Dynamic I/O Configuration Planning and Administration, SC24-6044.
Alternatively, using VM CP commands, you can add, delete, or modify the I/O definitions of channel
paths, control units, and I/O devices in the CPC. Using the IOCP utility on CMS (DYN option), you can
write a new POR IOCDS containing the same I/O configuration changes. For more information, see the
z/VM Planning and Administration, GA22-6974, and z/VM CP Command and Utility Reference, GA22-6974.
Overview of the z/VSE version of IOCP

SE The z/VSE version of IOCP runs under control of z/VSE. Run the z/VSE version of IOCP by using
z/VSE JCL statements. By coding options in the PARM parameter of the EXEC statement, you can:

IOCDSIOCP reads, validates, and checks the syntax of the input statements. It generates an IOCDS
based on the configuration defined in the statements. It produces reports of the IOCDS generated, and,
when requested, IOCP writes the generated IOCDS to the Support Element hard disk.
IOCP reads one or more IOCDSs into storage from the Support Element and prints the IOCP
Note: If you run VSE as a guest under VM, writing an IOCDS is not supported.
For a full description of the VSE version of IOCP, see Chapter 5, “Running the z/VSE version of IOCP,”
on page 131.
IOCP and system generation of a z/VSE operating system

You must define your I/O configuration to both software and hardware. VSE builds the ASI Procedure
file, which contains the statements that define the I/O configuration to your z/VSE operating system.
To define the I/O configuration to hardware, use a set of I/O statements in a separate IOCP file. IOCP
uses the I/O statements in the IOCP file to define the I/O configuration to the CPC.
Figure 15 shows an overview of the processing performed by the z/VSE version of IOCP.
1 4
2
ASI Procedure Input Statements IOCP Reports
IOCP
IOCDS in
Storage
5
VSE I/O
Configuration
IOCDSs
To HSA
Power-On Reset
1. For VSE system generation, the ASI Procedure file defines the I/O configuration information to the system
control program.
2. The input file contains the IOCP statements used to define the I/O configuration to IOCP.
3. IOCP processes the IOCP statements, generates an IOCDS, produces reports, and, when requested, writes the
5. The IOCDSs reside on the Support Element hard disk. At power-on reset, the Support Element loads the I/O
configuration data into the hardware system area (HSA). (The I/O subsystem uses the data in the hardware
system area (HSA) to control I/O requests.)
Figure 15. Overview of z/VSE version of IOCP

Overview of the stand-alone version of IOCP
Start and control the stand-alone version of IOCP from the CPC console. The CPC console is a logical
console available from the Hardware Management Console through the CPC Console task or from the
Support Element (SE).
Run the stand-alone version of IOCP to define I/O configuration data when you operate your system
before installing z/OS, VM, or z/VSE, or when you operate your system with a control program other
than these programs.
Start IOCP by selecting the I/O Configuration task from the Support Element or Hardware Management
Console. Briefly, the functions you can request are as follows:
1. Import a card-image statement file from the Hardware Management Console diskette drive (if it is
available), from a USB flash drive, or from an FTP location.
2. Export a card-image statement file to the Hardware Management Console diskette drive (if it is
available) or to a USB flash drive or to an FTP location.
3. Edit IOCP card-image data. You can supply card-image data by using the disassemble or import
functions, or the stand-alone IOCP editor.
4. Build an IOCDS read from the I/O configuration source (IOCS) file on the SE hard disk. IOCP writes
the IOCDS to the SE after a successful build.
5. Disassemble an IOCDS from the SE hard disk to create an IOCS file on the SE hard disk.
For a description of the stand-alone version of IOCP, see Chapter 6, “Running the stand-alone version of
IOCP,” on page 139. Figure 16 on page 51 shows an overview of the processing performed by the
stand-alone version of IOCP.

1 4
Service
Console IOCP Reports
IOCP
IOCDS in
2 Storage
5
Input Statements IOCDSs
IOCSs
To HSA
Power-On Reset
1. You control IOCP from the Support Element or Hardware Management Console.
2. The card-image input contains IOCP statements from the Support Element hard disk. You can also load card
images into the stand-alone IOCP editor from an I/O configuration source (IOCS) file.
3. IOCP processes IOCP window commands from the CPC console to control IOCP functions. If wanted, you can
use the stand-alone editor to inspect, create, or modify card images from an IOCS file using the CPC console.
Use the CPC console to:
v Import and export card image statements files
v Disassemble an IOCDS
v Build an IOCDS
v Print configuration reports
4. Configuration reports allow you to check the I/O configuration data as defined in the IOCDSs.
5. Build a new IOCDS from an IOCS file, or disassemble an existing IOCDS to generate an IOCS file. The
IOCDSs and IOCSs reside on the Support Element hard disk. When you perform power-on reset, the IOCDS is
loaded into the hardware system area (HSA). (The I/O subsystem uses the data in the HSA to control I/O
requests.)
Figure 16. Overview of stand-alone version of IOCP

Chapter 2. IOCP statements and functions
This chapter describes the following:
v Purpose of IOCP statements
v How to code them:
– Using z/OS
– Using z/VM
– Using z/VSE
Using this chapter

This chapter describes IOCP statements as you use them in z/OS, z/VM, and z/VSE operating systems.
It takes special care to identify any information that is specific to an operating system. This information is
identified as follows:
For z/OS: Information placed in this box is specific to the z/OS operating system
Specifying IOCP statements

The rules for coding IOCP card-image statements follow the rules for assembly language. See
Appendix A, “Coding IOCP statements,” on page 237 for a summary of the rules for coding statements
and the notation used in this document to illustrate statements.
IOCP has seven statements: ID, RESOURCE, UUID, CHPID, CNTLUNIT, IODEVICE, and FUNCTION.
The ID statement is optional; code it if you want specific identification information in the heading of
IOCP configuration reports and on Support Element (SE) displays. The RESOURCE statement is required
and defines the logical partitions and logical channel subsystems in your configuration. The UUID
statement is optional; code it if you want to enable UUID checking for functions in a logical partition.
CHPID, CNTLUNIT, IODEVICE, and FUNCTION statements are optional. CHPID defines channel paths,
CNTLUNIT defines control units, and IODEVICE defines I/O devices in your configuration. FUNCTION
defines the function an adapter will perform in your configuration for some adapters that do not require
a channel path.
Note: While IOCP requires only the RESOURCE statement, most I/O configurations need CHPID,
CNTLUNIT, and IODEVICE statements to be usable.
The following general rules apply to the sequence of IOCP statements in the input file:
v You can code the ID statement only once, and it must precede other IOCP statements.
v You can code the RESOURCE statement only once, and it must precede all CHPID, FUNCTION, and
UUID statements.
v You can code the UUID statement only once, and it must precede all FUNCTION statements.
v A CHPID statement that defines a channel path must precede a CNTLUNIT statement that refers to the
channel path.
v A CNTLUNIT statement must follow the CHPID statement that defines the channel path to which the
control unit is attached.
v An IODEVICE statement must follow the CNTLUNIT statement that defines the control unit to which
the device is assigned.
v You can insert IOCP comment cards (specified by an asterisk (*) in column 1 or by .* in columns 1 and
2) in the input file where you want them.

Because of these rules, it is suggested that you group all CHPID statements, followed by all CNTLUNIT
statements, and then all IODEVICE statements. Alternatively, you might group all CHPID statements,
followed by groups of related CNTLUNIT and IODEVICE statements (as shown in Appendix B, “Lists of
statement input,” on page 241).
IOCP and z/OS I/O configuration

The Hardware Configuration Definition (HCD) defines the I/O configuration to the z/OS operating
system and the I/O subsystem. You do not need to run IOCP directly to define the I/O configuration to
the I/O subsystem. However, HCD does run IOCP when it needs to write an IOCDS.
Figure 17 shows a typical input file containing IOCP statements.Appendix B, “Lists of statement input,”
on page 241 shows a list of a typical input file containing IOCP statements.
5
IODEVICE
4
CNTLUNIT
3
CHPID
2
RESOURCE
1 IOCP
ID Statements
1. ID statement (optional) defines heading data, or titles, for configuration reports. Only one ID statement is
allowed.
2. RESOURCE statement (required) defines the logical partitions and logical channel subsystems. Only one
RESOURCE statement is allowed.
3. CHPID statements (optional) define channel paths.
4. CNTLUNIT statements (optional) define control units. Each control unit must be defined in a separate
CNTLUNIT statement.
5. IODEVICE statements (optional) define I/O devices. As many as 256 I/O devices can be defined in an
IODEVICE statement.
6. UUID statement (optional and not pictured) specifies which logical partitions require UUID values.
7. FUNCTION statements (optional and not pictured) define functions.
Figure 17. IOCP statements
IOCP and z/VM I/O configuration

The information in some VM I/O configuration statements must correspond to the information in IOCP
statements (see Figure 18 on page 55). This section covers detailed information about the correspondence
between VM I/O configuration statements and IOCP statements.
Do not mix IOCP statements and VM I/O configuration statements in the same file. The IOCP source file
must be separate from the VM I/O configuration files. The VM I/O configuration files are the system
configuration file (SYSTEM CONFIG) and the obsolete real I/O configuration file (HCPRIO). The VM I/O

configuration files need to define only those devices that VM cannot sense during initial program load
(IPL) of VM or those devices for which you need to specify definition parameters (for example, shared
DASD).
As an alternative, the Hardware Configuration Manager (HCM) and the Hardware Configuration
Definition (HCD) define the I/O configuration to the VM operating system and the I/O subsystem. You
do not need to run IOCP directly to define the I/O configuration to the I/O subsystem. However, HCD
does run IOCP when it needs to write an IOCDS.
5
RDEVICE IODEVICE
4
CNTLUNIT
3
CHPID
VM I/O 2
Configuration RESOURCE
1
ID
IOCP Statements
allowed.
3. CHPID statements (optional) define channel paths. A VM I/O definition does not use a corresponding
statement.
CNTLUNIT statement. A VM I/O definition does not use a corresponding statement.
5. IODEVICE statements (optional) define I/O devices; a maximum of 256 I/O devices can be defined in an
IODEVICE statement. The information in the IODEVICE statements corresponds to the information in the
RDEVICE statements in the SYSTEM CONFIG file or the RDEVICE statements in the obsolete HCPRIO file.
Figure 18. Correspondence between IOCP statements and SYSTEM CONFIG statements or HCPRIO statements
IOCP and z/VSE I/O configuration

The information in z/VSE IPL ADD statements must correspond to information in IOCP IODEVICE
statements (see Figure 19 on page 56). Appendix B, “Lists of statement input,” on page 241 has sample
IOCP statements and the corresponding z/VSE IPL ADD statements.
Chapter 2. IOCP statements and functions 55

5
IPL ADD IODEVICE
STATEMENTS
4
CNTLUNIT
3
CHPID
2
RESOURCE
1
ID
IOCP Statements
allowed.
3. CHPID statements (optional) define channel paths. A z/VSE I/O configuration does not use a corresponding
statement.
CNTLUNIT statement. A z/VSE I/O configuration does not use a corresponding statement.
5. IODEVICE statements (optional) define I/O devices. A maximum of 256 I/O devices can be defined in an
IODEVICE statement. Information in the IODEVICE statements must correspond to information in the z/VSE
IPL ADD statements.
Figure 19. Correspondence between IOCP statements and z/VSE IPL ADD statements
IOCP definitions
The following list defines IOCP terms to which you can refer to when defining your I/O configuration:
Access list
Defines the name of one or more logical partitions (LPs) that have the channel path configured
online at LP activation after initial power-on reset of an IOCDS.
Candidate list
Defines the name of the logical partition or logical partitions that can configure the channel path
after LP activation.
Device candidate list
Defines the name of the logical partition or logical partitions that can access the device after LP
activation.
Director
Is an ESCON or FICON switch that provides a control unit port (CUP), which is accessed
through link address X'FE'.
ESCON channel path
IOCP defines the following channel path types (TYPE keyword) as ESCON channel paths:
v ESCON channel (CNC)
v ESCON channel that supports channel-to-channel communications (CTC)

v FICON channel attached to an ESCON Director (FCV).
ESCON control unit
A control unit that attaches to CNC, CTC, or FCV channel paths (TYPE keyword). An ESCON
control unit can also attach to FC channel paths.
ESCON CTC control unit
An ESCON CTC control unit is one of the following:
v A control unit that attaches to only TYPE=CTC channel paths
v A control unit that attaches to only TYPE=CNC or TYPE=FCV channel paths and has
UNIT=SCTC coded.
FICON CTC control unit
A FICON CTC control unit is a control unit that attaches to only TYPE=FC channel paths and has
UNIT=FCTC coded.
ESCON device
A device assigned to CNC, CTC, or FCV channel paths (TYPE keyword).
FICON channel path
A TYPE=FC channel path.
FICON director
A fibre channel switch that has a control unit port (CUP) and supports FC channel paths (TYPE
keyword).
FICON control unit
A control unit that attaches to only FC channel paths (TYPE keyword).
FICON device
A device assigned to only FC channel paths (TYPE keyword).
Parallel channel path
IOCP defines the following channel path types (TYPE keyword) as parallel channel paths:
v ESCON channel (CVC) that attaches to a 9034 ESCON Converter Model 1 or similar converter
and operates as a block multiplexer channel.
v ESCON channel (CBY) that attaches to a 9034 ESCON Converter Model 1 or similar converter
and operates as a byte multiplexer channel.
Parallel control unit
A control unit that attaches to CBY or CVC channel paths (TYPE keyword).
Parallel device
A device assigned to CBY or CVC channel paths (TYPE keyword).
Spanned channel path
IOCP defines a spanned channel path as a channel path that can be accessed by LPs in more than
one logical channel subsystem (CSS) at the same time. To define a spanned channel path, specify
more than one CSS ID in the CSS parameter of the PATH keyword on the CHPID statement. A
spanned channel path is also a shared channel path.
Shared channel path
IOCP defines a shared channel path as a channel path that can be accessed by more than one LP
in the same logical channel subsystem (CSS) at the same time. To define a shared channel path,
do one of the following actions:
v Specify the SHARED, NOTPART, or IOCLUSTER keyword on the CHPID statement
v Specify more than one logical partition in the access list of the PARTITION keyword on the
CHPID statement.
v Define the channel path as spanned by specifying more than one CSS ID in the CSS parameter
of the PATH keyword on the CHPID statement.

Reconfigurable channel path
IOCP defines a reconfigurable channel path as an unshared channel path that can be dynamically
moved between logical partitions. To define a reconfigurable channel path, specify the REC
parameter in the PARTITION keyword on the CHPID statement.
Dedicated channel path
IOCP defines a dedicated channel path as a channel path that can be accessed by only one logical
partition and cannot be moved between logical partitions. A dedicated channel path is unshared
and non-reconfigurable. To define a dedicated channel path, do not specify SHARED, NOTPART,
or REC in the CHPID statement and do one of the following actions:
v Specify a single logical partition in the access list of the PARTITION keyword in the CHPID
statement
v Specify a null access list and a single logical partition in the candidate list of the PARTITION
keyword in the CHPID statement.
ID
ID is an optional statement that can specify the following information:
v Identification information printed in the heading of IOCP configuration reports on the ID1 and ID2
lines. See Chapter 7, “IOCP configuration reports,” on page 151 for examples of the headings used on
the reports.
v Data set name for each IOCDS that appears on the Input/Output Configuration window of the
Support Element for a CPC
v Set of machine limits and rules to be enforced by IOCP when processing the I/O configuration
v Name of the local system.
When specified, the ID statement must precede all RESOURCE, CHPID, CNTLUNIT, IODEVICE, UUID,
and FUNCTION statements in the input stream. Also, when specified, it must include at least one
keyword (for example, MSG1, MSG2, or SYSTEM). The ID statement can be specified only once in the
input file. The ID statement parameters can appear in any order but you must code at least one
parameter. If you do not specify the ID statement, no identifying information is printed on the ID1 and
ID2 lines of the report headings.
See Appendix A, “Coding IOCP statements,” on page 237 for a summary of the rules for coding
statements and the notation used in this document to illustrate statements.
The format of the ID statement is as follows:

[symbol] ID {MSG1=’message’}
{MSG2=’message’}
{SYSTEM=(processor[,number])}
{LSYSTEM=name}
{TOK=token}
MSG1=
Specifies the identification information that is printed on the ID1 line of the heading on IOCP
configuration reports. The Support Element also uses the first eight characters of MSG1= (bytes 1-8)
as the customer name for the IOCDS that appears on the Input/Output Configuration window of the
Support Element for CPCs.
MSG2=
Specifies the identification information that is printed on the ID2 line of the heading on IOCP
configuration reports.
'message'
Specifies a string of 1-64 characters of identification information. IOCP accepts any character in the
string. Any incorrect character is translated to a quotation mark. See the “English (U.S.) I/O Interface
Code” table for EBCDIC characters in 3270 IDS Character Set Reference for a list of valid printable

characters. You must enclose the string within single quotation marks. Two consecutive single
quotation marks are counted as one character and indicate a single quotation mark within the string
(such as, MSG1=’John’’s I/O Report’).
SYSTEM=
Specifies the machine limits and rules that IOCP enforces for verification of the IOCP input or when
writing an IOCDS in preparation of a CPC upgrade. The stand-alone version of IOCP ignores the
SYSTEM keyword. It is also ignored when IOCP is writing an IOCDS using keyword
CHECKCPC=YES in the PARM parameter of the EXEC statement (z/OS or z/VSE) or CMS option
CHECKCPC. In both cases, IOCP uses the machine limits and rules for the CPC on which IOCP is
currently running.
When IOCP is writing an IOCDS in preparation of a CPC upgrade (keyword CHECKCPC=NO or
option NOCHKCPC is specified), IOCP uses the machine limits and rules based on the SYSTEM
keyword. IOCP requires the SYSTEM keyword when CHECKCPC=NO or NOCHKCPC is specified.
If the SYSTEM keyword is not specified and ICP IOCP runs on a CPC that it supports, IOCP uses the
machine limits for that CPC. If the SYSTEM keyword is not specified and ICP IOCP runs on an
unsupported CPC, IOCP uses default machine limits. See Table 16 on page 285 for the default
SYSTEM value.
See Appendix F, “Machine limits and rules,” on page 285 for a complete list of valid SYSTEM
keyword values and their associated machine limits and rules.
processor
Specifies the machine type number of the target CPC for which the IOCP input is to be verified. For
example, specify SYSTEM=(2084) if the IOCP input you need verified is for a 2084 CPC.
number
Specifies the machine limits number associated with the target CPC machine type for which the IOCP
input is to be verified. Specify a single decimal number. The default value is the lowest number
supported for the specified CPC machine type. For example, the default machine limits number for
the 2084 is one.
LSYSTEM=name
Specifies the name of the local system. The name is a 1-8 alphanumeric (0-9, A-Z) name. IOCP
requires the LSYSTEM keyword when CIB, CL5, and CS5 channel paths are defined. The CPC uses
the local system name to identify itself when establishing a coupling facility connection using CIB
channel paths. It is helpful to use the same CPC name specified on the Hardware Management
Console.
TOK=token
Configuration programs use this keyword to pass to the CPC information required to enable the
dynamic I/O configuration capability of any resulting IOCDS. This keyword is not intended for
direct IOCP user input. It is mentioned here to inform you that IOCP accepts the keyword and the
generated value.
Some CPCs have a fixed storage size for the Hardware System Area (HSA). These CPCs require every
IOCDS to have a token. If the TOK keyword is not specified, IOCP generates a token of its own and,
even though the Hardware Management Console might indicate dynamic I/O is enabled, no
operating system supports dynamic I/O configuration. See Table 16 on page 285 to determine if your
CPC has fixed HSA.
For HCD: Do not alter an IOCP input file generated by HCD and use it to write an IOCDS. If
changes are necessary, use HCD to regenerate the IOCP input.
ID Statement coding example

The following statement defines the identification information printed on lines ID1 and ID2 on the
heading of IOCP configuration reports. TPCONFIG is your name for the IOCDS that appears on the

Input/Output Configuration window of the Support Element. You have a 2084 Model A08 CPC and have
selected the appropriate machine limits and rules for your CPC with the SYSTEM keyword.
ID00B ID MSG1=’TPCONFIG Configuration for Processor’, X
MSG2=’Revised by Plan 3’, X
SYSTEM=(2084,1)
RESOURCE
RESOURCE is a required statement that defines:
v Logical partition names
v Multiple Image Facility (MIF) image ID numbers assigned to the logical partitions
v Logical channel subsystems (CSSs)
v The maximum number of devices that can be dynamically defined for a dynamic-capable IOCDS
v Reserved logical partitions, for a dynamic-capable IOCDS.
The RESOURCE statement must:

v Appear only once in an IOCP input file
v Follow the ID statement (if specified)
v Precede all CHPID, CNTLUNIT, IODEVICE, UUID, and FUNCTION statements
v Include every logical partition in the configuration
v Have either the PART or PARTITION parameter specified.
If the RESOURCE statement is not specified or is in error, IOCP terminates all processing and does not
validate the remaining IOCP statements.
The format of the RESOURCE statement is as follows:

[symbol] RESOURCE PART|PARTITION=((CSS(cssid),{name|({name|*}
[,MIF image ID])},...),...)
[MAXDEV=((CSS(cssid),ss[,ss][,ss]),...)]
PART=
PARTITION=
Specifies the logical partitions (LPs) in the configuration. You can use either form.
cssid
Specifies the logical channel subsystem (CSS) ID to define. The CSS ID is a 1-digit number in the
range 0-n. See Table 18 on page 287 to determine the maximum CSS ID your CPC supports.
All the logical partitions following a CSS ID parameter are defined in the CSS. You can define as
many CSSs as your CPC supports.
If you are defining only CSS 0, the CSS parameter is optional. Also, if you are defining only a single
CSS, the PART or PARTITION parameter can be enclosed in a single set of parentheses. For example:
PART=(CSS(1),LP1)
For a CPC with a fixed HSA storage size, IOCP defines any missing CSS so that every CSS your CPC
supports is defined. See Table 16 on page 285 to determine if your CPC has fixed HSA.
name
Specifies the name of the logical partition (LP) to be defined. The LP name is a 1-8 alphanumeric (0-9,
A-Z) character name that must have a nonnumeric first character.
The words PHYSICAL, REC, SYSTEM, and PRIMnnnn (where nnnn is a 4-digit decimal number) are
reserved and cannot be used as the names of LPs.

You must specify all LPs in the configuration. You can define up to 15 LPs in a CSS. Each LP name in
the configuration must be unique. See Table 18 on page 287 to determine the maximum number of
LPs your CPC supports.
A logical partition specified on the RESOURCE statement can exist without any channel paths and
functions assigned to it.
For z/OS: For coupling facility logical partitions (LPs), match the LP name specified in the PART or
PARTITION keyword of the RESOURCE statement with the NAME keyword of the CF statement in
the policy information defined for the coupling facility resource management (CFRM) policy of the
site.
* Specifies reserving an LP in the specified CSS with the specified MIF image ID. A reserved LP can be
specified only for a dynamic-capable IOCDS (TOK keyword on ID statement). A dynamic-capable
IOCDS is built when using HCD on OS/390, z/OS, or VM or when specifying IOCP CMS utility
option DYN for VM. A dynamic-capable IOCDS is also built for CPCs that have a fixed HSA storage
size. Space in the HSA is allocated for reserved LPs but cannot be used until a dynamic I/O
configuration change is made to assign a name and other resources to the LP.
The following rules apply when specifying reserved LPs:
v A reserved LP must have a user-specified MIF image ID.
v A reserved LP cannot have any channel paths assigned to it.
v An IOCDS cannot contain only reserved LPs.
v At least one LP must be defined with a name.
For a CPC with fixed HSA, IOCP defines enough reserved LPs to ensure that the maximum number
of LPs that your CPC supports is available in every CSS. See Table 16 on page 285 to determine if
your CPC has fixed HSA.
MIF image ID
Specifies the MIF (multiple image facility) image ID associated with a logical partition (LP). The MIF
image ID is a 1-digit hexadecimal number in the range 1-F (see note 9 on page 287). You can specify
the MIF image IDs non-consecutively in any order. MIF image IDs must be unique within a CSS but
can be duplicated in different CSSs.
Shared channel paths used for ESCON CTC or FICON CTC communication require a MIF image ID.
See the CUADD keyword “CUADD=address” on page 86 for more information.
If you do not specify a MIF image ID for every LP, IOCP arbitrarily assigns MIF image IDs to the
unassigned LPs.
The MIF image ID is different from the logical partition identifier (ID) displayed on the General page
of the image profile for CPCs.
For z/OS: For coupling facility logical partitions (LPs), the partition ID specified on the activation
profile for the CF image on the Support Element must match the number specified in the PARTITION
keyword of the CF statement in the policy information defined for the coupling facility resource
management (CFRM) policy of the site. It is helpful to use that the partition ID specified is a
concatenation of the CSS ID and MIF image ID. The MIF image ID is no longer used for this purpose.
MAXDEV=
Specifies the maximum number of devices, including those devices defined in the IOCDS, that can be
dynamically added to the configuration. IOCP ignores the MAXDEV keyword for CPCs that have a
fixed HSA storage size. For a CPC with fixed HSA, the maximum number of devices that are
supported by your CPC are available. See Table 16 on page 285 to determine if your CPC has fixed
HSA.
If the MAXDEV keyword is specified, the IOCDS must be dynamic-capable (TOK keyword on ID
statement). A dynamic-capable IOCDS is built when using HCD on OS/390, z/OS, or VM or
specifying IOCP CMS utility option DYN for VM.

If the MAXDEV keyword is not specified for a dynamic-capable IOCDS, IOCP defaults the maximum
number of devices to the maximum the CPC supports for subchannel set 0 for each defined CSS. If
you define devices in subchannel set 1 of any CSS, IOCP sets the default for the maximum number of
devices for each subchannel set 1 with devices to the maximum the CPC supports. Otherwise,
subchannel set 1 has a default value of zero.
If the MAXDEV keyword is specified for a dynamic-capable IOCDS but omits a defined CSS, IOCP
sets the default for the maximum number of devices to the maximum the CPC supports for
subchannel set 0 for the CSS. If you define devices in subchannel set 1 of the omitted CSS, IOCP sets
the default for the maximum number of devices for subchannel set 1 to the maximum the CPC
supports. Otherwise, subchannel set 1 has a default value of zero.
cssid
Specifies the logical channel subsystem (CSS) ID associated with this parameter. The CSS ID is a
1-digit number in the range of 0-n. See Table 18 on page 287 to determine the maximum CSS ID your
CPC supports.
The CSS ID must have been defined in the PART or PARTITION keyword on the RESOURCE
statement. You can specify as many CSSs as you defined. If only a single CSS was defined, the CSS
parameter is optional. Also, if you only specify a single CSS, the MAXDEV parameter can be enclosed
in a single set of parentheses. For example:
MAXDEV=(CSS(1),60K)
ss Specifies the maximum number of devices, including those devices defined in the IOCDS, that can be
used for dynamic I/O configuration in a subchannel set in a CSS. The first ss value defines the
maximum number of devices for subchannel set 0 in a CSS. The second ss value defines the
maximum number of devices for subchannel set 1 in the same CSS. See Table 18 on page 287 to
determine the number of subchannel sets your CPC supports. All CPCs support subchannel set 0.
To specify a value for subchannel set 1, you must also specify a value for subchannel set 0.
The format of ss is:
{number | MIN | MAX}
number
Specifies the maximum number of devices, including those devices defined in the IOCDS, that
can be used for dynamic I/O configuration in a subchannel set in a CSS. The number can be a 1-
to 5-digit decimal number in the range 0-n. The n is the maximum number of devices your CPC
supports for the subchannel set. The number can also be in the form nnK. The nn is a 1- or
2-digit decimal number that is multiplied by 1024 to get the maximum number of devices. (For
example, 60K and 61440 are equivalent values.) See Table 18 on page 287 to determine the
maximum number of devices per subchannel set that your CPC supports. The configuration
cannot contain more devices than the MAXDEV value.
MIN
Specifies that the maximum device value in the I/O configuration is equal to the number of
devices defined in the IOCDS for the subchannel set in the CSS. Therefore, no new devices can be
dynamically added to the I/O configuration for the subchannel set in the CSS unless other
devices have been deleted.
MAX
Specifies that the maximum device value in the I/O configuration for the subchannel set in the
CSS is equal to the maximum number of devices supported by your CPC for the subchannel set
(see Table 18 on page 287).

RESOURCE statement coding examples
Example 1
RESOURCE PART=((LP2,2),(LP1,1),(LP3,3))
The preceding statement:

v Defines logical channel subsystem (CSS) 0
v Defines three logical partitions in CSS 0: LP1, LP2, and LP3
v Assigns MIF image ID number 1 to LP1, MIF image ID number 2 to LP2, and MIF image ID number 3
to LP3.
Example 2
RESOURCE PART=(CSS(1),(LP1,2),LP2,(LP3,5))

v Defines CSS 1
v Defines three logical partitions in CSS 1: LP1, LP2, and LP3
v Assigns MIF image ID number 2 to LP1, MIF image ID number 5 to LP3, and IOCP arbitrarily assigns
a MIF image ID number to LP2.
Example 3
RESOURCE PART=((CSS(0),(LP1,1),LP2),(CSS(1),(LPA,1),(*,5),(LPB,F))),
MAXDEV=((CSS(0),40K,12K),(CSS(1),56820))

v Defines CSSs 0 and 1
v Defines two logical partitions in CSS 0: LP1 and LP2
v Defines two logical partitions and reserves space for one other logical partition in CSS 1: LPA and LPB
are defined and one logical partition is reserved until a future dynamic I/O change assigns a name to
the logical partition.
v Assigns MIF image ID number 1 to LP1 and LPA, MIF image ID number F to LPB, MIF image ID
number 5 to the reserved logical partition in CSS 1, and IOCP arbitrarily assigns a MIF image ID
number to LP2.
v Specifies that the maximum number of devices that can be defined, including dynamic I/O additions,
in the configuration for CSS 0 is 40960 for subchannel set 0 and 12288 for subchannel set 1 and for CSS
1 is 56820 for subchannel set 0. Assuming that there are no devices defined in subchannel set 1 in CSS
1, the maximum number of dynamic I/O additions for subchannel set 1 is zero–that is, no dynamic
I/O additions are allowed.
CHPID
CHPID is an optional statement that defines:
v Channel path identifiers (chpids)
v Physical channel identifiers (pchids)
v Channel path types
v chpid assignment to ESCON and FICON Directors
v Connecting chpids
v Channel path parameter
v I/O cluster name owning a chpid

v Adapter identifiers (AIDs), ports, and connecting system names
v Availability of chpids to logical partitions
v Reconfigurability of chpids among logical partitions
v Shared chpids for logical partitions
v Spanned chpids for logical channel subsystems
v Physical network identifier.
When specified, the CHPID statement must:

v Follow the ID statement (if specified) and the RESOURCE statement
v Precede any CNTLUNIT statements that refer to the chpid it defines
v Collectively, specify all chpids to use in the configuration.
Channel paths not specified in a CHPID statement cannot be used.
statements and the notation in this document to illustrate statements.
The format of the CHPID statement is as follows:

[symbol] CHPID PATH=(CSS(cssid[,cssid]...),chpid number)
[PCHID=number]
[VCHID=number]
TYPE={CBY | CVC | CNC | CTC | CFP | CFR | CFS |
CBP | CBR | CBS | ICP | CIB | CL5 | CS5 | OSC | OSD |
OSE | OSM | OSN | OSX | FC | FCV | FCP | IQD }
[MIXTYPE]
[SWITCH=number]
[PART | PARTITION={(CSS(cssid),{name|0}[,REC])|
(CSS(cssid),access list)|
(CSS(cssid),(access list)[,(candidate list)][,REC])|
((CSS(cssid),(access list)[,(candidate list)]),...)}]
[NOTPART={(CSS(cssid),access list)|
((CSS(cssid),(access list)[,(candidate list)]),...)}]
[SHARED]
[PNETID=([,...]physical network id[,...])
[CPATH=(CSS(cssid),chpid number)]
[CHPARM=value]
[IOCLUSTER=name]
[AID=aid number]
[PORT=number]
[CSYSTEM=name]
PATH=
cssid
Specifies the logical channel subsystem (CSS) associated with this parameter. A CSS ID is a 1-digit
number in the range 0-n. See Table 18 on page 287 to determine the maximum CSS ID your CPC
supports.
The CSS IDs must have been defined in the PART or PARTITION keyword on the RESOURCE
statement. You can specify as many CSSs as you defined. If only a single CSS was defined, the
CSS parameter is optional.
If you specify multiple CSS IDs for a channel path, IOCP defines the channel path as spanned. A
spanned channel path is available for use by the logical partitions in each specified CSS. All
channel paths can be spanned except CNC, CTC, CVC, CBY, FCV, CFR, and CBR (TYPE
keyword). See Table 23 on page 294 to determine which channel path types your CPC supports
being spanned.
chpid number
Specifies a 2-digit hexadecimal channel path ID (chpid) number for the channel path for each CSS

ID specified in the CSS parameter. A chpid cannot be duplicated within a CSS. See Appendix F,
“Machine limits and rules,” on page 285 to determine the valid chpid ranges your CPC supports.
PCHID=number
Specifies the physical channel identification (pchid) number associated with the channel path. The
pchid number is a 1- to 3-digit hexadecimal number in the range 0-n where n is the maximum number
your CPC supports. See Appendix F, “Machine limits and rules,” on page 285 to determine the
maximum pchid number your CPC supports.
The pchid number identifies the physical location (cage, slot, card port) for a channel path in the CPC.
The CPC maintains this mapping. This number can change, for example in ESCON port sparing–if
one card port fails, the CPC enables the spare port and assigns the pchid number to the new port.
PCHIDs are generally unique to the card slot, some slots have multiple PCHIDs, some only one. A
PCHID on a CHPID statement must not match a PCHID on a FUNCTION statement or the reverse.
The PCHID keyword is not allowed for internal channel paths ICP and IQD (TYPE keyword) and for
CIB and CS5 channel paths. The PCHID keyword is required for all other channel path types. PCHID
and VCHID keywords are mutually exclusive and only one can be used on a given CHPID statement.
Multiple CL5 channel paths can be specified with the same pchid. You can define a maximum of 8
CL5 channel paths with the same pchid.
The keyword is optional when performing a deck verification. It is required when writing an IOCDS.
IOCP can validate an IOCP input file without pchids by performing a deck verification (WRTCDS=NO
or NOWRTCDS). You can see the PCHIDs available for the I/O configuration of the CPC on the
PCHID Report. (See “Using a PCHID report to help plan your I/O configuration” on page 17.) You
can either use the PCHID Report and assign pchid numbers to channel paths or use the CHPID
Mapping Tool. The CHPID Mapping Tool is available from Resource Link at www.ibm.com/servers/
resourcelink. The tool can use the IOCP input file and add the appropriate pchid numbers for your
configuration to the CHPID statements. IOCP can then use the input file to write an IOCDS.
Note: An IOCP input file that HCD created without pchids must be migrated back into HCD after the
CHPID Mapping Tool has added pchid numbers to the file. HCD can then write an IOCDS, or a new
IOCP input file can then be created from HCD that stand-alone IOCP can use to write an IOCDS.
VCHID=number
Specifies the virtual channel identification (vchid) number associated with the channel path. The vchid
number is a 1- to 3-digit hexadecimal number in the range n-m where n is the minimum number and
m is the maximum your CPC supports. See Table 18 on page 287 to determine the minimum and
maximum vchid numbers your CPC supports. If no vchid numbers are listed, then the VCHID
keyword is not supported on that particular machine.
The vchid number identifies the (virtual) channel path in the CPC. The CPC maintains this mapping.
The VCHID keyword is allowed only for channel path TYPE=IQD (internal queued direct
communication (HiperSockets)). The VCHID keyword is required for TYPE=IQD channel paths on
CPCs that have a minimum and maximum vchid.
TYPE=
Specifies the mode of I/O operation for the channel path. See Appendix F, “Machine limits and
rules,” on page 285 to determine which channel path types your CPC supports.
CBY
Specifies that the channel path is an ESCON channel that attaches to a 9034 ESCON Converter
Model 1 or similar converter. IOCP defines CBY channel paths as parallel channel paths.
CBY channel paths operate the same as parallel byte multiplexer channel paths. They operate in
burst mode or byte-interleave mode, depending on the attached control unit.
Notes:
1. CBY channel paths require ESCON channel hardware.

2. If you define an ESCON channel as CBY but do not attach it to an ESCON Converter, the
channel is put into a permanent standby error state at power-on reset (POR). Put the channel
in single channel service mode before POR to avoid the channel error. However, the channel
is still unusable until it is attached to an ESCON Converter.
CVC
Specifies that the channel path is an ESCON channel that attaches to a 9034 ESCON Converter
Model 1 or similar converter. IOCP defines CVC channel paths as parallel channel paths.
CVC channel paths operate the same as parallel block multiplexer channel paths. They operate
only in burst mode and allow multiplexing between blocks. See Figure 43 on page 261 for CVC
channel path coding examples.
Notes:
1. CVC channel paths require ESCON channel hardware.
2. If you define an ESCON channel as CVC but do not attach it to an ESCON Converter, the
channel is put into a permanent standby error state at POR. Put the channel in single channel
service mode before POR to avoid the channel error. However, the channel is still unusable
until it is attached to an ESCON Converter.
CNC
Specifies that the channel path is an ESCON channel, and that all control units and I/O devices
that attach to this channel path support the ESCON Architecture protocol. See Appendix E,
“Configuration examples,” on page 253 for CNC channel path coding examples.
CTC
Specifies that the channel path is an ESCON channel that permits channel-to-channel
communications. See Appendix E, “Configuration examples,” on page 253 for CTC channel path
coding examples.
A CTC channel path communicates with CNC channel paths and, conversely, a CNC channel
path communicates with CTC channel paths. A CTC channel path cannot communicate with
another CTC channel path, nor can a CNC channel path communicate with another CNC channel
path.
Notes:
1. You must specify identical unit addresses in the CNTLUNIT statements for CTC and CNC
channel paths that communicate with each other.
2. You must specify the MIF image ID or logical partition numbers associated with shared CNC
and CTC channel paths in the CUADD keyword on the CNTLUNIT statements for the CTC
communication. See ESCON and FICON Channel-to-Channel Reference, SB10-7034, for more
information.
3. When specifying a control unit associated with an unshared CNC and CTC channel path, do
not use the CUADD keyword.
CFP
Specifies that the channel path is a coupling facility peer channel. CFP channel paths are defined
for ISC-3 links. Each CFP channel path assigned to a control unit should have seven devices
defined.
A CFP channel path connects to another CFP channel path.
CFR
Specifies that the channel path is a coupling facility receiver channel. CFR channel paths are
defined for ISC-3 links. You can define a CFR channel path only for a coupling facility logical
partition to use.
A CFR channel path connects to a CFS channel path. A CFR channel path cannot connect to
another CFR channel path.

CFS
Specifies that the channel path is a coupling facility sender channel. CFS channel paths are
defined for ISC-3 links. Each CFS channel path should have two devices defined.
A CFS channel path connects to a CFR channel path. A CFS channel path cannot connect to
another CFS channel path.
CBP
Specifies that the channel path is an Integrated Cluster Bus coupling facility peer channel. CBP
channel paths are defined for ICB-3 and ICB-4 links. Each CBP channel path assigned to a control
unit should have seven devices defined.
A CBP channel path connects to another CBP channel path. However, an ICB-3 link must connect
to another ICB-3 link and an ICB-4 link must connect to another ICB-4 link.
CBR
Specifies that the channel path is an Integrated Cluster Bus coupling facility receiver channel.
CBR channel paths are defined for ICB-2 links. You can define a CBR channel path only for a
coupling facility logical partition to use.
A CBR channel path connects to a CBS channel path. A CBR channel path cannot connect to
another CBR channel path.
CBS
Specifies that the channel path is an Integrated Cluster Bus coupling facility sender channel. CBS
channel paths are defined for ICB-2 links. Each CBS channel path should have two devices
defined.
A CBS channel path connects to a CBR channel path. A CBS channel path cannot connect to
another CBS channel path.
ICP
Specifies that the channel path is an Internal Coupling facility peer channel. Each ICP channel
path assigned to a control unit should have seven devices defined.
An ICP channel path connects to another ICP channel path. You must connect every ICP channel
path to another ICP channel path by specifying the CPATH keyword in the CHPID statement.
You cannot connect an ICP channel path to itself.
See “Defining internal coupling channels (TYPE=ICP)” on page 30 for recommendations on
defining ICP channel paths.
CIB
Specifies that the channel path is a coupling over InfiniBand channel. Each CIB channel path
assigned to a control unit should have seven devices defined.
A CIB channel path connects to another CIB channel path. You must connect every CIB channel
path to another CIB channel path by specifying the CPATH and CSYSTEM keywords in the
CHPID statement. You cannot connect a CIB channel path to itself. If a CIB channel path is
defined, the LSYSTEM keyword on the ID statement is required.
See “Defining coupling over InfiniBand or PCIe-O (TYPE=CIB, TYPE=CL5, or TYPE CS5)
channels” on page 30 for more information about defining CIB channel paths.
CL5
Specifies that the channel path is a Coupling Express LR channel. Each CL5 channel path
assigned to a control unit should have 32 devices defined but IOCP tests for a minimum of eight.
A CL5 channel path connects to another CL5 channel path. You must connect every CL5 channel
path to another CL5 channel path by specifying the CPATH and CSYSTEM keywords in the
CHPID statement. You cannot connect a CL5 channel path to itself. If a CL5 channel path is
defined, the LSYSTEM keyword on the ID statement is required

channels” on page 30 for more information about defining CL5 channel paths.
CS5
Specifies that the channel path is a Coupling Short Reach channel (Integrated Coupling Adapter
(ICA SR)). Each CS5 channel path assigned to a control unit should have eight devices defined.
A CS5 channel path connects to another CS5 channel path. You must connect every CS5 channel
path to another CS5 channel path by specifying the CPATH and CSYSTEM keywords in the
CHPID statement. You cannot connect a CS5 channel path to itself. If a CS5 channel path is
defined, the LSYSTEM keyword on the ID statement is required.
channels” on page 30 for more information about defining CS5 channel paths.
OSC
Specifies that the channel path is an open systems adapter (OSA) channel that operates as an
OSA-Express integrated console controller (OSA-ICC). An OSC channel path provides 3270
sessions that can serve as the system operator or master console for an operating system so you
can IPL. An OSC channel path can also connect to TPF through 3215 data streams.
OSA-ICC is supported only on OSA-Express, OSA-Express2, and OSA-Express3 1000Base-T
Ethernet cards. If an OSC channel path is specified on any other type of channel hardware, the
I/O subsystem flags the channel path with a definition error and the channel path is unavailable
until a dynamic I/O change is made or power-on reset is performed with a new IOCDS to
resolve the error.
For more OSC channel path configuration information, see Open Systems Adapter-Express Integrated
Console Controller User's Guide, SA22-7990.
Note: Devices assigned to OSC channel paths must be configured on the Support Element before
use.
OSD
Specifies that the channel path is an open systems adapter (OSA) channel supporting the Queued
Direct I/O (QDIO) architecture. An OSD channel path supports IP applications providing
connectivity to Gigabit Ethernet, 10 Gigabit Ethernet, 1000Base-T Ethernet, Fast Ethernet, and
Token Ring LANs. All OSA channel hardware supports OSD channel paths.
For more OSD channel path configuration information, see Open Systems Adapter-Express
Customer's Guide and Reference, SA22-7935.
OSE
Specifies that the channel path is an open systems adapter (OSA) channel that uses non-QDIO
architectures. An OSE channel path provides connectivity for all LAN protocols supported by the
non-QDIO architecture (1000Base-T Ethernet, Fast Ethernet, and Token Ring LANs). OSE channel
paths are supported on OSA-Express and OSA-Express2 features. All OSA channel hardware
except Gigabit Ethernet and 10 Gigabit Ethernet support OSE channel paths.
For more OSE channel path configuration information, see Open Systems Adapter-Express
OSM
Specifies that the channel path is an open systems adapter (OSA) channel providing the interface
for the zBX management services. The OSM channel path provides access to the intranode
management network (INMN). Only OSA-Express3 or later 1000Base-T Ethernet can support the
OSM CHPID type.
For more OSM channel path configuration information, see Open Systems Adapter-Express

OSN
Specifies that the channel path is an open systems adapter (OSA) channel for network control
program (NCP). An OSN channel path supports traffic between an operating system image (for
example, z/OS) and a Communication Controller for Linux (CCL) image. OSN channel paths are
supported on OSA-Express2 and OSA-Express3 features Gigabit Ethernet and 1000Base-T
Ethernet.
Note: OSA-Express-4S Gigabit Ethernet features do not support the OSN CHPID type.
For more OSN channel path configuration information, see Open Systems Adapter-Express
OSX
Specifies that the channel path is an open systems adapter (OSA) channel providing the interface
for software to access data on the zBX blades. The OSX channel path provides access to the
intraensemble data network (IEDN). Only OSA-Express3 or later 10 Gigabit Ethernet can support
the OSX CHPID type.
For more OSX channel path configuration information, see Open Systems Adapter-Express
FC Specifies that the channel path is a FICON channel and that all control units and I/O devices that
attach to this channel path support the FICON Architecture protocol. All FICON channel
hardware supports FC channel paths.
FC channel paths can communicate with each other in channel-to-channel communications. See
Appendix E, “Configuration examples,” on page 253 for an FC channel-to-channel coding
example.
Notes on defining FICON CTC communication:

1. You must specify a control unit type (UNIT keyword) of FCTC for FC channel paths to
communicate with each other.
2. You must specify identical unit addresses in the CNTLUNIT statements for FC channel paths
that communicate with each other.
3. You must specify the CUADD keyword on the CNTLUNIT statements for CTC
communication when the remote channel path is a shared FC channel path. If the remote
channel path is on a CPC that does not support multiple logical channel subsystems, then the
logical address specified must equal the partition number of the logical partition with which
you want to communicate. If the remote channel path is on a CPC that does support multiple
logical channel subsystems, then the logical address specified must equal the combination of
the CSS ID and the MIF image ID of the logical partition with which you want to
communicate. (For example, specify CUADD=15 for MIF image ID 5 in CSS 1.) For additional
information, see the ESCON and FICON Channel-to-Channel Reference, SB10-7034.
4. When specifying an FCTC control unit associated with an unshared FC channel path, do not
use the CUADD keyword.
FCV
Specifies that the channel path is a FICON channel that attaches to an ESCON Director. IOCP
defines FCV channel paths as ESCON channel paths. FCV channel paths operate the same way as
ESCON channel paths.
Notes:
1. FCV channel paths require FICON Express LX channel hardware. If an FCV channel path is
specified on any other type of channel hardware, including FICON Express2 LX, the I/O
subsystem flags the channel path with a definition error and the channel path is unavailable
until a dynamic I/O change is made or power-on reset is performed with a new IOCDS to
resolve the error.

2. If you define a FICON channel as FCV but do not attach it to an ESCON Director, the channel
is put into a permanent error state after POR. Put the channel in single channel service mode
before POR to avoid the channel error. However, the channel is still unusable until it is
attached to an ESCON Director.
FCP
Specifies that the channel path is a FICON channel that operates as a Fibre Channel Protocol
channel. All FICON channel hardware supports FCP channel paths. See “Fibre Channel Protocol
support” on page 37 for more information about FCP channel paths.
IQD
Specifies that the channel path is an internal queued direct communication (HiperSockets)
channel. See “Defining IQD channels” on page 37 for recommendations on defining IQD channel
paths.
MIXTYPE=
Specifies that FC and FCP channel path types can be defined on the same FICON adapter. Beginning
with the FICON Express16S+ adapter, the two ports on the adapter must both be defined as either FC
or FCP. If you define a mixture of channel path types on the adapter, one channel path will be usable
while the other channel path will be placed in definition error and be unusable. CPCs that support
the FICON Express 16S+ adapter may also have prior FICON adapters that do not have this
restriction against mixing channel path types. On these CPCs, IOCP defaults to preventing the mixing
of FC and FCP channel path types on an adapter. So if the CPC has any FICON adapters prior to
FICON Express 16S+ for which you want to define FC and FCP channels on the same adapter, the
MIXTYPE keyword must be specified for at least one of the CHPID statements to permit the channel
path type mixture. To determine if your CPC requires the MIXTYPE keyword to mix FC and FCP
channel path types for an adapter, see “Machine rules” on page 285.
The MIXTYPE keyword is valid only for FC and FCP channel paths. Also, the MIXTYPE keyword is
ignored if specified for a CPC that does not require the keyword to allow a channel path type
mixture.
IOCP determines that channel paths are on the same adapter when their PCHIDs are in the same
modulo-4 PCHID slot.
SWITCH=number
Specifies an arbitrary number for the IBM 9032 or 9033 ESCON Director or a FICON Director to
which the channel path is assigned. The number can range from 00-FF for a maximum of 256 for each
IOCDS. A channel path can be assigned to only one number. A FICON Director cannot have the same
switch number as an ESCON Director.
For an ESCON channel path using multiple ESCON Directors, specify the switch number for the
director that has dynamic connections for the channel path. For a FICON channel path, always
specify the switch number of the entry switch. The entry switch is the switch to which the FICON
channel actually connects. The switch number for a FICON Director is not the same as the switch
address which is set up by using switch control panels. Whenever possible, try to use the same value
for switch numbers and switch addresses to reduce the number of ways a director is known. For
more information about defining FICON Directors, see “Defining FICON switch configurations” on
page 38.
The SWITCH keyword is required as follows:
v For channel paths to have dynamic connections through a director
v For FCV channel paths (TYPE keyword)
v For managed channel paths (CHPARM keyword with, for example, a value of 01).
The SWITCH keyword is valid only for CBY, CVC, CNC, CTC, FCV, and FC channel paths (TYPE
keyword). For ESCON Directors, the SWITCH keyword is valid for CBY, CVC, CNC, CTC, and FCV
channel paths. However, CBY and CVC channel paths can have only dedicated connections through

an ESCON Director. For FICON Directors, the SWITCH keyword is valid for FC channel paths. See
Appendix E, “Configuration examples,” on page 253 for SWITCH keyword coding examples.
In either of the following cases, the I/O subsystem flags the channel path with a definition error:
v If you specify the SWITCH keyword for an FC channel path (TYPE keyword) that does not attach
to a FICON Director
v If you do not specify the SWITCH keyword for an FC channel path that does attach to a FICON
Director.
The channel path is unavailable until it is attached to or detached from a FICON Director to match
the configuration definition and it is varied off and on, or until a dynamic I/O change is made or
power-on reset is performed with a new IOCDS to resolve the error.
PART=
PARTITION=
Specifies the availability of chpids to logical partitions. All logical partition names and CSS IDs that
you specify must match the ones specified in the RESOURCE statement.
You must specify the PART, PARTITION, NOTPART, SHARED, or IOCLUSTER keyword in every
CHPID statement or define the channel path as spanned (PATH keyword). You can specify the PART
or PARTITION keyword and the SHARED or NOTPART keywords in the same CHPID statement.
However, you cannot specify the PART or PARTITION and NOTPART keywords for the same CSS ID
in the same CHPID statement.
Either form of this keyword can be used in a CHPID statement. However, only one form of this
keyword can be in each statement.
cssid
1-digit number in the range 0-n. See Table 18 on page 287 to determine the maximum CSS ID
your CPC supports.
Each CSS ID must have been specified in the PATH keyword for this channel path. You can
specify as many CSSs as you specified in the PATH keyword. If only a single CSS ID was
specified in the PATH keyword, the CSS parameter is optional. If the channel path is spanned
and more than one CSS was specified in the PATH keyword, the CSS parameter is required.
name
Specifies the name of a logical partition (LP) that can access the channel path. The LP name is a
1-8 alphanumeric (0-9, A-Z) character name that must have a nonnumeric first character and must
have been defined on the RESOURCE statement. You cannot specify a reserved logical partition
(*).
0 Specifies that no LPs can access the channel path following LP activation for the initial POR of
the IOCDS. The zero indicates a null access list.
If you specify a zero without parentheses, you cannot specify a candidate list and you must do
one of the following actions:
v Specify the SHARED keyword for a shared channel path
v Specify multiple CSS IDs in the PATH keyword for a spanned channel path
v Specify the REC parameter for a reconfigurable channel path.
access list
Specifies the logical partition (LP) or logical partitions (LPs) in the CSS that have the channel
path configured online at LP activation following initial power-on reset (POR) of an IOCDS.
After the initial POR of an IOCDS, PR/SM LPAR retains which LPs have the channel path
configured online at LP activation following subsequent PORs of the IOCDS. In this case, the
access list used to build an IOCDS might not match the access list PR/SM LPAR retains.

Logical partitions that are not specified in the access list of a shared, spanned, or reconfigurable
channel path can access the channel path after POR through channel reconfiguration commands if
the LP is specified in the channel candidate list of the path. (See “candidate list.”
The format of the access list is:
{name[,name]...|0}
name
See “name” on page 71.
0 Specifies that no LPs in the CSS accesses the channel path following LP activation for the
initial POR of the IOCDS. The zero indicates a null access list.
The following rules apply when specifying the access list:
v If you specify more than one LP name, IOCP defines the channel path as shared. All channel
paths can be shared except CBY, CVC, CFR, and CBR (TYPE keyword).
v If you specify a zero for a null access list in a CSS, you cannot specify a logical partition name
for the same CSS.
v If you specify either one LP name or a zero (a zero indicates a null access list), the channel
path can be shared, reconfigurable, or dedicated (unshared and non-reconfigurable).
v If you specify a zero as the access list (null access list), you must do one of the following
actions:
– Specify the SHARED keyword for a shared channel path
– Specify multiple CSS IDs in the PATH keyword for a spanned channel path
– Specify the REC parameter for a reconfigurable channel path
– Specify a candidate list consisting of one LP name for a dedicated channel path.
v An LP name can appear only once in an access list.
candidate list
Specifies the logical partitions (LPs) in the CSS that can access the channel path. LPs that are not
specified in a channel candidate list of the path cannot access the channel path.
If you specify a candidate list, you do not need to repeat the LP names specified in the access list.
IOCP automatically includes in the candidate list the LP names specified in the access list.
The format of the candidate list is:
{name[,name]...|=}
name
See “name” on page 71.
= Specifies that the candidate list for the CSS is identical to its access list.
The following rules apply when specifying the candidate list:

v You cannot specify an equal sign (=) for a candidate list of the channel path if the access list is
zero (0 indicates a null access list).
v If you do not specify a candidate list for a CSS to which the channel path is assigned and the
channel path is shared, spanned, or reconfigurable, the candidate list defaults to all logical
partitions defined in each CSS to which the channel path is assigned, excluding reserved
logical partitions (*).
v If you do not specify a candidate list and the channel path is dedicated, the candidate list
defaults to the logical partition specified in the access list.
v If a dedicated channel path has a zero specified as the access list (null access list), you must
specify a candidate list of exactly one LP name.
v If the candidate list consists of more than one LP name, the channel path must either be
shared, spanned, or reconfigurable.

v An LP name can appear only once in a candidate list.
REC
Specifies that the channel path is reconfigurable. Specifying a channel path as reconfigurable
allows it to be dynamically moved between logical partitions (LPs) in a CSS. To move it you use
channel reconfiguration commands after power-on reset (POR) is complete.
The following rules apply when specifying the REC parameter for a channel path:
v The channel path must have an access list of one LP name or zero (0 indicates a null access
list).
v If a candidate list is specified, it must consist of one or more LP names.
v If the candidate list is not specified, it defaults to all LPs defined in the CSS excluding reserved
LPs (*).
v You cannot specify:
– The REC parameter and the SHARED keyword
– The REC parameter and an access list of more than one LP
– The REC parameter for a spanned channel path
– An access list of one LP and a candidate list with an LP other than the one in the access list,
unless you specify either the REC parameter or the SHARED keyword or define the channel
path as spanned.
– A zero as the access list (null access list) and a candidate list of two or more LPs, unless you
specify either the REC parameter or the SHARED keyword or define the channel path as
spanned.
– A zero as the access list (null access list) and no candidate list, unless you specify either the
REC parameter or the SHARED keyword or define the channel path as spanned.
NOTPART=
Specifies the availability of chpids to logical partitions. All logical partition names and CSS IDs that
you specify must match the ones specified in the RESOURCE statement.
You must specify the NOTPART, PART, PARTITION, SHARED, or IOCLUSTER keyword in every
CHPID statement or define the channel path as spanned (PATH keyword). You can specify the
NOTPART keyword and the SHARED, PART, or PARTITION keywords in the same CHPID
statement. However, you cannot specify the NOTPART and PART or PARTITION keywords for the
same CSS ID in the same CHPID statement.
The NOTPART keyword indicates that the channel path is shared. All channel paths can be shared
except CBY, CVC, CFR, and CBR (TYPE keyword).
cssid
your CPC supports. If the channel path is spanned and more than one CSS ID was specified in
the PATH keyword, the CSS parameter is required.
Each CSS ID must have been specified in the PATH keyword for this channel path. You can
specify as many CSSs as you specified in the PATH keyword. If only a single CSS was specified
in the PATH keyword, the CSS parameter is optional.
access list
Specifies the logical partitions or partitions in a CSS that do not have the channel path configured
online at LP activation after initial power-on reset (POR) of an IOCDS.
IOCP generates an access list for the channel path consisting of all LPs in the CSS not specified in
the NOTPART access list. For example, if a CSS has three LPs (LP1, LP2, and LP3) and you
specify NOTPART=LP2, IOCP generates an access list containing LP1 and LP3.

After the initial POR of an IOCDS, PR/SM LPAR retains which logical partitions have the
channel path configured online at LP activation after subsequent PORs of the IOCDS. In this case,
the access list used to build an IOCDS might not necessarily match the access list PR/SM LPAR
retains.
Logical partitions that are not specified in the access list of a shared, spanned, or reconfigurable
channel path can access the channel path after POR through channel reconfiguration commands if
the LP is specified in the candidate list of the channel path. (See “candidate list.”)
{name[,name]...|0}
name
Specifies the name of a logical partition (LP) that cannot access the channel path. The LP
name is a 1-8 alphanumeric (0-9, A-Z) character name. It must have a nonnumeric first
character. It must have been defined on the RESOURCE statement. You cannot specify a
reserved logical partition (*).
0 Specifies that no LPs in the CSS access the channel path following LP activation for the initial
POR of the IOCDS. A zero indicates a null access list.
v If you specify all LP names in a CSS or a zero, no LPs in the CSS access the channel path
following LP activation for the initial POR of the IOCDS. (IOCP generates a null access
list.)
v If you specify a zero for a null access list in a CSS, you cannot specify a logical partition
name for the same CSS.
v If you specify a NOTPART candidate list, you do not need to repeat in the NOTPART
access list the LP names specified in the NOTPART candidate list. IOCP automatically
excludes from the generated access list the LP names specified in the NOTPART candidate
list.
For example, if a CSS has three LPs (LP1, LP2, and LP3) and you specify
NOTPART=((LP2),(LP3)), IOCP generates an access list containing LP1.
v An LP name can appear only once in an access list.
candidate list
Specifies the logical partitions (LPs) in the CSS that cannot access the channel path.
IOCP generates a candidate list for the channel path consisting of all LPs in the CSS not specified
in the NOTPART candidate list. For example, if a CSS has three LPs (LP1, LP2, and LP3) and you
specify NOTPART=((LP2),(LP3)), IOCP generates a candidate list containing LP1 and LP2.
LPs that are not in the candidate list IOCP generates cannot access the channel path.
{name[,name]...|=}
name
See “name.”
= Specifies that the candidate list of the channel path is identical to its access list.
The following rules apply when specifying the NOTPART candidate list:
v You cannot specify an equal sign (=) for the candidate list of a channel path if the access list is
zero (0 indicates a null access list).
v You cannot specify all LP names in a CSS because doing so generates a null candidate list for
the CSS.
v If you do not specify a candidate list for a CSS to which the channel path is assigned for the
NOTPART keyword, the candidate list IOCP generates defaults to all LPs defined in the CSS or
CSSs, excluding reserved logical partitions (*).

SHARED
Specifies that the channel path on the CHPID statement is shared. All channel paths can be
shared except CBY, CVC, CFR, and CBR (TYPE keyword).
You must specify the SHARED, PART, PARTITION, NOTPART, or IOCLUSTER keyword in every
CHPID statement or define the channel path as spanned (PATH keyword).
A shared channel path can be accessed by more than one logical partition (LP) at the same time.
Unshared channel paths can be accessed only by a single LP. Although a reconfigurable channel
path can be dynamically moved between LPs, only a single LP can access it at any given time.
The following rules apply to shared channel paths:
v If a channel path has an access list of more than one LP or the NOTPART keyword is specified
for it, the channel path defaults to shared.
v A spanned channel path (PATH keyword) defaults to shared.
v If you specify SHARED without the PART or NOTPART keywords, the access and candidate
lists default to all LPs defined in the CSSs to which the channel path is assigned, excluding
reserved logical partitions (*).
v You cannot specify:
– SHARED for reconfigurable channel paths
– An access list of one LP and a candidate list with an LP (other than the LP in the access list
with the PART or PARTITION keyword), without specifying either the REC parameter, the
SHARED keyword, or defining the channel path as spanned
– A zero as the access list (null access list) and a candidate list of two or more LPs with the
PART or PARTITION keyword, without specifying either the REC parameter, the SHARED
keyword, or defining the channel path as spanned
– A zero as the access list (null access list) and no candidate list with the PART or PARTITION
keyword, without specifying either the REC parameter, the SHARED keyword, or defining
the channel path as spanned.
PNETID=(physical-network-id[,...])
Specifies the physical network identifiers (pnetids) associated with the ports on the channel
adapter. Each pnetid is a 1-16 alphanumeric (0-9, A-Z) value.
The pnetid in the PNETID keyword is a positional parameter. One to four pnetids can be specified
which correspond to the ports on the channel adapter. You can omit one or more pnetids by
specifying just a comma for that adapter port position in the keyword value. For example, to only
specify the second pnetid for a channel path to be associated with the second port of the adapter,
you would specify PNETID=(,pnetid ).
Physical network IDs are used for channel path types OSD and IQD to establish the affinity to a
particular physical network. When network connections require two interfaces to be coordinated,
as with the 10GB RoCE network function adapter and an associated OSD channel path, the
pnetids should match. The network adapters that support the OSD channel path type have up to
four ports so four pnetids are supported.
An IQD channel path only uses a pnetid associated with the first port. For IQD channel path
types, all PNETID values must be unique.
CPATH=
Specifies the ICP, CIB, CL5, or CS5 channel path to which the channel path being defined is
connected.
The CPATH keyword is valid only for ICP, CIB, CL5, and CS5 channel paths (TYPE keyword)
and is required for all ICP, CIB, CL5, and CS5 definitions. It specifies the channel path that
connects to the ICP, CIB, CL5, or CS5 channel path being defined.

For example, PATH=(FE),CPATH=(FF),TYPE=ICP,... specifies that ICP channel path FE is connected
to channel path FF. ICP channel paths are internal channel paths and therefore the connected
channel paths exist on the same CPC. CIB, CL5, and CS5 channel paths connect to a CIB, CL5, or
CS5 channel path, respectively, on the same CPC or a different CPC.
cssid
Specifies the logical channel subsystem (CSS) ID associated with this parameter. The CSS ID
is a 1-digit number in the range 0-n. See Table 18 on page 287 to determine the maximum
CSS ID the connecting CPC supports.
The CSS IDs must have been defined in the PART or PARTITION keyword on the
RESOURCE statement of the CPC with the connecting channel path. For both internal and
external connections, specify one CSS ID associated with the connecting channel path. If the
connecting channel path is spanned, only one CSS ID is specified in this parameter.
If the channel path being defined is spanned, the CSS parameter is required. If only a single
CSS ID was specified in the PATH keyword and the connecting channel path has the same
CSS ID, the CSS parameter is optional.
chpid number
Specifies a 2-digit hexadecimal number for the connecting chpid number.
The CPATH keyword has extra syntax rules when the connected channel paths belong to the
same CPC (that is, they are internal connections). All ICP connections are internal. CIB, CL5,
and CS5 connections are internal when the CSYSTEM keyword value matches the LSYSTEM
keyword value on the ID statement for the CPC. The following syntax rules apply when
specifying the CPATH keyword for internal connections:
v The channel path specified in the CPATH keyword does not need to be defined before its
use in CPATH but must be defined somewhere in the configuration.
v Channel paths cannot connect to each other if they both have candidate lists with the same,
single logical partition. This restriction prevents the definition of ICP and internally
connected CIB, CL5, or CS5 channels in a configuration with only one logical partition.
v An ICP, CIB, CL5, CS5 channel path must connect to another channel path of the same
type and can connect to only one path.
v A channel path cannot connect to itself.
v Connecting two CIB or CS5 channel paths that have the same AID and port is not
supported. Also, connecting two CL5 channel paths that have the same PCHID and port is
not supported. IOCP does not allow these types of configuration.
CHPARM=value
Specifies a parameter for the channel path that indicates how the channel is to operate. The value is a
2-digit hexadecimal number.
The CHPARM keyword identifies a channel path to be managed by dynamic CHPID management.
When bit 7 of the CHPARM value is on (for example, X'01'), the channel path is managed. Only
CNC, FCV, and FC channel paths (TYPE keyword) can be managed. A managed channel path cannot
be spanned (PATH keyword). Managed paths require that the IOCDS to be dynamic-capable (TOK
keyword). A dynamic-capable IOCDS is built when using HCD on OS/390, z/OS, or VM or
specifying IOCP CMS utility option DYN for VM.
The CHPARM keyword also specifies whether an OSD or OSX channel path is to disable priority
specification. When bit 6 of the CHPARM keyword value is on (for example, X'02'), the OSD or OSX
channel path has priority specification disabled and the channel supports a maximum of 1920 OSA
valid subchannels (640 TCP/IP stacks). Otherwise, the OSD or OSX channel path has priority
specification enabled, and the channel supports a maximum of 480 OSA valid subchannels (160
TCP/IP stacks). OSM channel paths default to disabled priority queuing, and that is the only mode
supported for OSM CHPIDs.

Disabling priority specification is supported only for OSA-Express2 or later features. If it is disabled
for an OSA-Express feature, the I/O subsystem flags the channel path with a definition error. The
channel path is unavailable until a dynamic I/O change is made or a power-on reset is performed
with a new IOCDS to resolve the error. For additional information, see “I/O configuration
considerations” on page 21.
The CHPARM keyword also specifies the TCP/IP maximum transmission unit (MTU) and the IQDIO
frame size to use for I/O requests on IQD channel paths (TYPE keyword). Bits 0 and 1 of the
CHPARM keyword value are used and support the following values for IQD channel paths:
v 00 for 8 K MTU and 16 K frame size
v C0 for 56 K MTU and 64 K frame size.
On 2817 and 2818 with enabling MCL applied or 2828, 2827, or later processors, the following
CHPARM values are provided for HiperSocket IQD channel paths (TYPE keyword):
v Cx for 56 KB MTU and 64 KB frame size
v x0 for a normal HiperSocket channel path
v x2 for a HiperSocket for IEDN channel path. You can specify a CHPARM value of x2 for only one
IQD channel path.
v x4 for a HiperSocket capable to be externally bridged channel path
The CHPARM keyword also specifies whether an OSC channel path is to connect to TPF through
3215 data streams. When the required RPQ is installed on your CPC and bit 1 of the CHPARM value
is on (for example, X'40'), the channel path can connect to TPF using 3215 data streams. Otherwise,
the OSC channel path supports TN3270E emulator sessions.
Any undefined value is incorrect for all channel path types.
The following table summarizes the supported CHPARM values:
Table 9. Supported CHPARM values
CHPID type
CHPARM
value CNC, FCV, FC OSC OSD, OSX OSM IQD All others
00 Not Managed TN3270E Priority Warning. 8 K MTU N/A
Enabled Defaults to
priority
disabled
01 Managed Error Error Error Error Error
02 Error Error Priority Priority 8 K MTU; Error
Disabled Disabled IEDN
04 Error Error Error Error 8 K MTU; ext Error
bridge
40 Error 3215 data Error Error 16 K MTU Error
streams
42 Error Error Error Error 16 K MTU; Error
IEDN
bridge
80 Error Error Error Error 32 K MTU Error
82 Error Error Error Error 32 K MTU; Error
IEDN

Table 9. Supported CHPARM values (continued)
CHPID type
CHPARM
value CNC, FCV, FC OSC OSD, OSX OSM IQD All others
bridge
C0 Error Error Error Error 56 K MTU Error
C2 Error Error Error Error 56 K MTU; Error
IEDN
C4 Error Error Error Error 56 K MTU; ext Error
bridge
IOCLUSTER=name
Specifies an I/O cluster name. An I/O cluster is a sysplex that owns a managed channel path
(CHPARM keyword). The I/O cluster name is a 1-8 alphanumeric (0-9, A-Z) or special (@, #, or $)
character name. The maximum number of I/O cluster names is 64 or 128. See “Machine limits” on
page 286 to determine the maximum number of I/O cluster names your CPC supports.
Only operating systems running in logical partitions (LPs) that belong to the specified sysplex can
manage the channel path.
The IOCLUSTER keyword must be specified for every managed channel path. The IOCLUSTER
keyword cannot be specified for any channel path that is not managed.
The PART, PARTITION, and NOTPART keywords cannot be specified on the CHPID statement when
the IOCLUSTER keyword is specified. Also, the IOCLUSTER keyword indicates that a channel path is
shared and assigns the channel path a null access list and a candidate list of all LPs defined in the
configuration.
AID=aid number
Specifies the adapter identifier (aid) associated with the host channel adapter (HCA) or Integrated
Coupling Adapter (ICA) on which this channel path is defined. The aid is a one- to 2-digit
hexadecimal number. See Table 18 on page 287 to determine the aid range your CPC supports.
The aid is determined from the PCHID Report when you order an HCA or ICA (see “Using a PCHID
report to help plan your I/O configuration” on page 17) and from Hardware Management Console
windows.
The AID keyword is valid only for CIB and CS5 channel paths (TYPE keyword) and is required for
all CIB and CS5 definitions. Multiple CIB or CS5 channel paths can be specified with the same aid.
You can define a maximum of 16 CIB channel paths or 8 CS5 channel paths with the same aid.
PORT=number
Specifies the port on the host channel adapter (HCA), Coupling Express (CE LR) adapter, or
Integrated Coupling Adapter (ICA) to which this channel path is defined. The port number is a
1-digit decimal number in the range 1-n where n is the maximum number of HCA, CE LR, or ICA
ports your CPC supports. See Table 17 on page 286 and Table 18 on page 287 to determine the
maximum port numbers your CPC supports.
The PORT keyword is valid only for CIB, CL5, and CS5 channel paths (TYPE keyword) and is
required for all CIB, CL5, and CS5 definitions. Multiple CIB and CS5 channel paths can be specified
for the same port on an AID. Multiple CL5 channel paths can be specified for the same port on a
PCHID.
CSYSTEM=name
Specifies the name of the system that connects to this channel path. The name is a 1-8 alphanumeric
(0-9, A-Z) name.
For a coupling facility connection to be established, the specified CSYSTEM keyword value must
match the LSYSTEM keyword value for the CPC with the target CIB, CL5, and CS5 channel path.

The CSYSTEM keyword is valid only for CIB, CL5, and CS5 channel paths (TYPE keyword) and is
required for all CIB definitions. The CSYSTEM keyword is optional for CL5 and CS5 definitions.
You cannot connect two CIB, CL5, or CS5 channel paths to the same destination channel path. That
is, you cannot specify the same CSYSTEM and CPATH keyword values for two CIB, CL5, or CS5
channel paths.
CHPID statement coding examples
Example 1
CHPID PATH=17,PCHID=100,TYPE=CNC,PARTITION=((LP2),(LP1,LP4),REC),SWITCH=3E

v Defines channel path 17 in the only CSS in the configuration for the card port associated with physical
channel ID 100
v Specifies that the channel path operates as an ESCON channel
v Specifies that logical partition LP2 initially accesses the channel path and that LP1 and LP4 are also
authorized to access the channel path
v Defines the chpid as reconfigurable
v Defines the chpid the switch number 3E.
Example 2
CHPID PATH=28,PCHID=100,TYPE=CNC,PART=((LP1,LP3),(LP2))

channel ID 100
v Specifies that the channel operates as an ESCON channel path
v Implicitly defines the channel path as shared
v Specifies that logical partitions LP1 and LP3 initially access the channel path and that LP2 is also
authorized to access the channel path.
Example 3
CHPID PATH=28,PCHID=100,TYPE=CNC,NOTPART=((LP1,LP3),(LP2))

channel ID 100
v Specifies that the channel operates as an ESCON channel path
v Specifies that all logical partitions in the configuration except LP1, LP2, and LP3 initially access the
channel path and that all logical partitions in the configuration except LP2 are also authorized to access
the channel path.
Example 4
CHPID PATH=35,PCHID=100,TYPE=CNC,PART=((0),(LP2,LP3)),SHARED

channel ID 100

v Specifies that the channel path operates as an ESCON channel
v Defines the channel path as shared
v Specifies that no logical partition initially accesses the channel path and that LP2 and LP3 are
authorized to access the channel path.
Example 5
CHPID PATH=(CSS(1),35),PCHID=100,TYPE=FC,PART=(LP1,LP2)

v Defines channel path 35 in CSS 1 for the card port associated with physical channel ID 100
v Specifies that the channel path operates as a FICON channel
v Specifies that logical partitions LP1 and LP2 in CSS 1 initially access the channel path. Any other
logical partitions in CSS1 in the configuration are also authorized to access the channel path.
Example 6
CHPID PATH=(CSS(0,1),FF),TYPE=IQD,PART=((CSS(0),(LP1,LP2),(=)),(CSS(1),
(LPA),(=)))

v Defines channel path FF in CSSs 0 and 1
v Specifies that the channel path is a HiperSockets channel
v Defines the channel path as spanned and implicitly as shared
v Specifies that only logical partitions LP1 and LP2 in CSS 0 and logical partition LPA in CSS 1 can
access the channel path.
Note: See Appendix E, “Configuration examples,” on page 253 for more CHPID statement coding
examples.
Example 7
CHPID PATH=(CSS(0),80),TYPE=FC,SHARED,PCHID=1A0
CHPID PATH=(CSS(2),86),TYPE=FCP,SHARED,PCHID=1A1,MIXTYPE
The preceding statements:

v Defines channel path 80 in CSS 0 for PCHID 1A0
v Defines channel path 86 in CSS 2 for PCHID 1A1
v Specifies that the mixture of CHPID types is allowed on this CPC for these two channel paths. The
MIXTYPE keyword could be specified on either or both of the CHPID statements. If the CPC does not
require the MIXTYPE keyword when mixing FC and FCP channel path types on an adapter, the
MIXTYPE keyword is ignored and a warning message results. If the FICON adapter does not support
the mixing of CHPID types, one of the channel paths will be in definition error and unusable.
CNTLUNIT
CNTLUNIT is an optional statement that describes the following:
v The characteristics of the control unit
v The channel paths to which the control unit is attached
v The unit addresses the control unit recognizes.

Each physical control unit in the I/O configuration must be specified in a separate CNTLUNIT statement
to be used. This requirement includes control units that reside in the same physical unit as an I/O device
or in another control unit.
You are recommended to define only one CNTLUNIT statement for each physical control unit, control
unit image (defined with CUADD values), or control-unit function in your configuration. However, there
are configurations that require you to use multiple CNTLUNIT statements:
v Some storage controls (for example, the IBM 3990) contain multiple control-unit functions with the
number varying based on the mode of the storage control (for example, DLS or DLSE in the 3990
model 3). You are recommended to code a CNTLUNIT statement for each control-unit function (storage
director) provided in the storage control.
v You might require multiple logical partitions (LPs) to access a single control unit, perhaps with
multiple paths per LP. In this case, try to use shared channel paths to a single control unit definition.
However, if doing so is not possible (for example, if the configuration calls for more than eight paths to
a parallel control unit), you can use multiple CNTLUNIT statements to define the single control unit
function. In this case, ensure that each LP does not have paths defined to more than one CNTLUNIT
definition.
v If your control unit supports both ESCON and parallel interfaces, you might want to use both types of
interfaces to the same LP. You are recommended to avoid this type of configuration, but if you require
the mixing of interfaces, you must consider the following rules:
– You must choose between all-parallel or all-ESCON interfaces to any one LP at any one time. You
can use one CNTLUNIT statement to define a control-unit image for the parallel paths and another
CNTLUNIT statement to define a control-unit image for the ESCON paths. Each CNTLUNIT must
be accompanied with unique IODEVICE statements.
– When using the ESCON paths to the control unit, you must:
- Vary offline all devices (that is, device numbers) defined to the parallel control-unit image.
- Disable the parallel interfaces at the physical control unit.
– When using the parallel paths, vary offline all devices (that is, device numbers) defined to the
ESCON control-unit image by doing one of the following tasks:
- If you have an ESCON Director, use the ESCON Manager (or the ESCON Director console) to
prohibit the connection between the control unit and the system
- If your ESCON connections are point-to-point, configure the chpids offline using the operating
system, ESCON Manager or the CPC console.
– When you are running VM with guest operating systems, you can use multiple CNTLUNIT
statements for a single physical control unit in certain environments to effectively dedicate the same
physical devices to more than one guest. This technique involves potential path-grouping
considerations that create operational complications. Ensure you have determined possible
consequences and that you use caution if employing this technique.
See Appendix F, “Machine limits and rules,” on page 285 for a list of machine limits and rules for your
CPC. See Appendix D, “List of I/O devices and control units,” on page 245 for a list of some control unit
types and their characteristics.
Based on information in the CNTLUNIT statements, IOCP constructs control unit headers for a logical
channel subsystem (CSS) to use. A control unit header is a logical representation of one to eight physical
control units. A control unit header is built for each one of the following units:
v Physical control unit with no I/O devices attached.
v Physical control unit that does not share any I/O devices with other physical control units.
v Groups of two or more control units that share devices between them.
A CSS uses the control unit headers for queuing I/O requests for the devices attached to the associated
physical control units. One I/O request queue exists for each control unit header. A CSS adds to the same

I/O request queue all I/O requests for all devices in the control unit header. (Devices in the control unit
header are all the devices attached to the set of physical control units that are in the control unit header.)
IOCP also establishes a rotation order for the channel paths associated with each control unit header.
When initiating I/O requests associated with the control unit header, a CSS uses this rotation order to
determine the sequence for selecting channel paths. See “Rotation order” on page 44 for a description of
how IOCP establishes the rotation order of the channel paths.
The CNTLUNIT statement must follow any CHPID statements that specify channel paths the control unit
uses. The CNTLUNIT statement must precede any IODEVICE statements that refer to the control unit.
The format of the CNTLUNIT statement is as follows:

[symbol] CNTLUNIT CUNUMBR=number
PATH=((CSS(cssid),chpid[,chpid]...),...)
[LINK=((CSS(cssid),link address
[,link address]...),...)]
[CUADD=address]
[PROTOCL={D|S|S4}]
SHARED={Y|N}
UNIT=type
UNITADD=((address[,number]),...)
CUNUMBR=number
Specifies the hexadecimal number that is assigned to the control unit. One to four hexadecimal digits
in the range of 0000-FFFE can be specified. You must assign a unique number to each control unit.
The numbers can be assigned arbitrarily.
PATH=
cssid
Specifies the logical channel subsystem (CSS) IDs associated with this parameter. A CSS ID is a
your CPC supports.
The CSS IDs must have been defined in the PART or PARTITION keyword on the RESOURCE
statement. You can specify as many CSSs as you defined. If only a single CSS was defined, the
CSS parameter is optional.
chpid
Specifies the channel paths in each CSS that are attached to the control unit. You must specify
two hexadecimal digits for each channel path identifier (chpid) or two asterisks (**). You cannot
duplicate chpid numbers within a CSS.
You can specify a maximum of eight chpids in a CSS. However, you can specify only one channel
path in a CSS for:
v An ESCON CTC control unit
v A FICON CTC control unit
v A control unit defined for FCP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel paths (TYPE
keyword).
You can specify a maximum of seven pairs of asterisks in a CSS. Each pair of asterisks specifies
that a managed path (CHPARM keyword) can be added to the control unit and identifies the
control unit as being managed by dynamic CHPID management. The number of pairs of asterisks
specifies the number of managed paths that can be added to the control unit. The following
restrictions apply to a managed control unit:
v At least one chpid number must be specified in each CSS

v You can specify a maximum of eight chpids and pairs of asterisks in a CSS
v Only CNC, FCV, and FC channel paths (TYPE keyword) can be attached to the control unit
v The channel paths attached to the control unit must be shared or spanned.
A spanned channel path is not required to attach to a control unit from each CSS to which the
channel path is assigned. The channel paths attaching to the control unit must be all unshared or
all shared. You cannot mix shared and unshared channel paths to a control unit. All spanned
channel paths are also shared channel paths.
You can assign a CNC, CTC, or FC channel path (TYPE keyword) to only one control unit unless
one of the following statements is true:
v The channel path is assigned to an ESCON or FICON Director (SWITCH keyword)
v The control unit has a logical address (CUADD keyword).
You can assign a CFS, CFP, CBS, CBP, ICP, CIB, CL5, and CS5 channel path (TYPE keyword) to
only one control unit. You cannot assign a CFR or CBR channel path (TYPE keyword) to a control
unit. CFR and CBR channel paths do not require control units or devices. You can assign an OSC,
OSE, OSN, or FCP channel path (TYPE keyword) to only one control unit.
You can assign an IQD, OSD, OSM, or OSX channel path (TYPE keyword) to only one control
unit unless the control unit has a logical address (CUADD keyword). If you need to define more
than 256 devices for an IQD channel path or 254 devices for an OSD, OSM, or OSX channel path,
define multiple control units. Specify a unique logical address for each control unit using the
CUADD keyword.
All channel path types (TYPE keyword) attaching to a control unit must meet the following
restrictions:
v All CBY or CVC
v All CNC, FCV, or FC
v CTC
v All CFS or CBS
v All CFP, CBP, ICP, CIB, CL5, or CS5
v OSC
v OSD
v OSE
v OSM
v OSN
v OSX
v FCP
v IQD
You can intermix FC channel paths with CNC and FCV channel paths. To ease migrating from
ESCON channels to FICON channels using dynamic I/O configuration, mix FC channel paths
with CNC and FCV channel paths. The mixture allows you to dynamically add FC channel paths
to a control unit while keeping its devices operational. A second dynamic I/O configuration
change can then remove the CNC and FCV channels while keeping the devices operational.
For some types of control units and devices, there might be small operational performance
differences when intermixing (for example, Resource Measurement Facility™ (RMF™) usage
calculations and channel path reconnection). Dynamic CHPID management cannot manage a
control unit that contains intermixing. For these control units and devices, mixing of FC channel
paths with CNC and FCV channel paths should be temporary. Non-dynamic I/O users are
recommended to avoid mixing these channel paths.
When mixing ICP, CIB, CL5, and CS5 channel paths in the same control unit, the CSYSTEM
keyword value on the CIB, CL5, and CS5 (if specified) channel paths must be identical to the

local system name of the CPC (LSYSTEM keyword). When mixing CIB, CL5, and CS5 channel
paths in the same control unit, any CSYSTEM keyword values that are specified for the channel
paths must be identical. If a CS5 or CL5 channel path did not specify a CSYSTEM keyword value,
it is ignored.
Other types of control units and devices fully support this intermixing. To determine the level to
which your control units and devices support intermixing, see their device planning manuals. For
control units that support intermixing, consult their performance white papers for guidance on
how to best configure channels from a performance perspective. To improve the balance of
workload across channels in an intermixed environment, alternate ESCON (CNC and FCV) and
FICON (FC) channel paths in the control unit header for this control unit. Use the following
patterns:
v For 6 ESCON and 2 FICON, use a sequence of E, E, E, F, E, E, E, F
v For 4 ESCON and 4 FICON, use a sequence of E, F, E, F, E, F, E, F
v For 2 ESCON and 6 FICON, use a sequence of E, F, F, F, E, F, F, F.
LINK=
cssid
Specifies the logical channel subsystem (CSS) IDs associated with this parameter. A CSS ID is a
your CPC supports.
Each CSS ID must have been specified in the PATH keyword for this control unit. You can specify
as many CSSs as you specified in the PATH keyword. If only a single CSS was specified in the
PATH keyword, the CSS parameter is optional.
link address
Specifies the link addresses to which the control unit is attached. The order in which the link
addresses are specified for each CSS corresponds to the order in which the channel paths are
specified in the CSS parameters of the PATH keyword. For example, the second link address
identified for CSS 1 in the LINK keyword corresponds to the second channel path identified for
CSS 1 in the PATH keyword.
The LINK keyword is optional if none of the channel paths in the PATH keyword attach to an
ESCON or FICON Director (SWITCH keyword). You do not need to specify link addresses for a
CSS if none of the channel paths specified for the same CSS ID in the PATH keyword attaches to
an ESCON or FICON Director. You can specify two or four asterisks for link addresses that
correspond to channel paths not attached to a director. For ESCON channel paths not attached to
a director, specify any valid link address values. FC channel paths not attaching to a director can
specify only link addresses of asterisks. A link address of asterisks must be specified for each
channel path in the PATH keyword that is specified as two asterisks (**). This pairing of asterisks
in the PATH and LINK keywords of the CNTLUNIT statement indicates the number of managed
paths (CHPARM keyword) that can be added to the control unit.
You must specify link addresses with two or four hexadecimal digits. The valid link address
range is 01-FE for CBY, CVC, CNC, CTC, and FCV channel paths (TYPE keyword). The valid link
address range is 0000-FFFF for an FC channel path. To determine whether to specify a 2- or
4-digit link address for an FC channel path and how to select a 4-digit link address, see “Defining
FICON switch configurations” on page 38. When you specify a 4-digit link address for an FC
channel path, every link address for that channel path must also be four digits. If you specify the
LINK keyword for a CSS, you must specify either a valid link address or asterisks for every
channel path in the same CSS in the PATH keyword. If you do not specify the LINK keyword or
specify asterisks, IOCP assumes that the link address is arbitrary and assigns a link address of
0D. Link addresses specified in the LINK keyword for dynamic connections through a director
must equal the link addresses associated with the physical ports on the director to which the
control unit is attached. However, link addresses specified for channel paths having dedicated
connections through an ESCON Director can be arbitrary.

For channel paths having dedicated connections through an ESCON Director, specify the
corresponding link address on the CNTLUNIT statement as the link address associated with the
physical port on the ESCON Director to which the control unit is attached. If a CTC channel path
does not attach to an ESCON Director or has a dedicated connection through an ESCON Director,
the link address specified here might not be the link address the CSS uses.
For maximum availability of a control unit and its devices, see Appendix D, “List of I/O devices
and control units,” on page 245 for recommendations for selecting the physical ports on an
ESCON Director to attach to the control unit.
The LINK keyword is valid only for CBY, CVC, CNC, CTC, FCV, and FC channel paths (TYPE
keyword).
Even though you can specify multiple link addresses for CBY and CVC channel paths (TYPE
keyword), the I/O subsystem does not support dynamic connections through an ESCON Director
for CBY or CVC channel paths. A CBY or CVC channel path that attaches to an ESCON Director
can have only a dedicated connection through it. Therefore, you can connect only one link
address on an ESCON Director to a CBY or CVC channel path at a time.
Note: If you specify no link address for a channel path that does not attach to a director, IOCP
selects a link address of 0D to address the control units on the channel path. This arbitrary link
address is displayed by the I/O Problem Determination (IOPD) frame and the IOCDS Display
frame on the Support Element. However, the IOCP reports show that no link address was
specified on its CNTLUNIT statements.
The I/O subsystem flags a channel path with a definition error in the following situations:
v The channel path does not attach to an ESCON or FICON Director or has a dedicated
connection through an ESCON Director and has more than one link address specified on its
CNTLUNIT statements.
v The channel path has a dynamic connection through a director and no link address was
specified on the CNTLUNIT statement.
v The channel path has 4-digit link addresses but is not connected to a high integrity fabric of
FICON switches. For more information about high integrity fabrics, see “Defining FICON
switch configurations” on page 38.
In these three instances, the channel path is unavailable and remains so until the error is
resolved.
In addition to the preceding syntax rules, the following configuration rules also apply when
specifying link addresses. For link addresses with four digits, rules are enforced against the
switch to which the channel path is connected (that is, the entry switch) rather than the switch to
which the control unit is attached. When a 4-digit link address is specified for a channel path,
specify four-digit link addresses for all paths that have the same entry switch. Otherwise, IOCP
cannot enforce some of the following switch configuration rules.
v Specifying more than one link address for an ESCON channel path is not allowed unless the
channel path is connected to an ESCON Director. Specifying any link address other than
asterisks for an FC channel path is not allowed unless the channel path is connected to a
FICON Director.
v All channel path types (TYPE keyword) assigned to a link address on an ESCON Director must
be identical, with the following exceptions:
– CBY or CVC channel paths
– CNC or FCV channel paths.
These channel paths are the only ones you can assign to the same link address on an ESCON
Director.
v Specification of any given path (pchid.link) on more than one control unit is not allowed unless
one of the following statements is true:

– The control units attach to CBY or CVC channel paths (TYPE keyword).
– The control units have unique logical addresses (CUADD keyword). See “Example 6” on
page 91 for CNTLUNIT definition considerations.
v The specification of FICON CTC and non-FICON CTC control units (see “IOCP definitions” on
page 56) to the same link address on the same director is not allowed. The control units must
either all specify UNIT=FCTC or none of them can specify FCTC.
v The specification of multiple control units to the same link address on the same director is not
allowed unless one of the following conditions is true:
– The control units attach to CBY or CVC channel paths (TYPE keyword).
– The control units are ESCON CTC or FICON CTC control units. (See “IOCP definitions” on
page 56.) The path (pchid.link.cuadd) attaching to each control unit is unique. That is, the
pchid, link address, or logical address (CUADD keyword) for the control unit must be
different on each control unit.
– The control units have the same unit address ranges and none of the paths attaching to the
different control units belongs to the same logical partition. The logical partition intersection
test is based on the initial access lists of the channel paths attaching to the control units.
– Unique logical addresses (CUADD keyword) were specified on all the CNTLUNIT
statements.
FC control unit definition considerations:
FC channel paths (TYPE keyword) support a maximum number of unit address ranges and unit
addresses. (See Table 17 on page 286.) IOCP calculates the number of unit address ranges and
unit addresses for a spanned FC channel path (PATH keyword in the CHPID statement) as
follows:
v If a single link address is used for the channel path in each CSS of a control unit, one unit
address range is counted and the number of unit addresses specified with the UNITADD
keyword on the CNTLUNIT statement are counted once.
v If different link addresses are used for the channel path in any of the CSSs of a control unit,
one unit address range is counted for each unique link address and the number of unit
addresses specified with the UNITADD keyword on the CNTLUNIT statement are counted
once for each unique link address.
For example, CHPID 10 is an FC channel path that is spanned to CSSs 0, 1, 2, and 3. It is
assigned to a control unit in CSSs 0, 1, and 3. The channel path uses link address C0 in CSSs 0
and 1 and link address D0 in CSS 3. Because there are a total of two unique link addresses, two
unit address ranges are counted. If the UNITADD keyword specified 128 unit addresses, a total of
256 unit addresses are counted.
CUADD=address
Specifies the logical address for the control unit.
For example, the IBM Enterprise Storage Server® (ESS) 2105 control unit requires that the logical
address specified with the CUADD keyword matches the control unit image address associated with
the corresponding control unit logical subsystem (LSS).
Note: Not all ESCON and FICON control units support logical addressing.
The address is 1-2 hexadecimal digits in the range 0-F for CNC, CTC, and FCV channel paths and in
the range 00-FF for FC, IQD, OSD, OSM, and OSX channel paths. However, if FC channel paths are
mixed with CNC or FCV channel paths, the address must be in the range 0-F. The address allows
you to specify a CNC, CTC, FCV, or FC channel path on multiple CNTLUNIT statements for a given
path (pchid.link or switch.link) or to specify an IQD, OSD, OSM, or OSX channel path on multiple
CNTLUNIT statements.
v For CNC, CTC, and FCV channel paths, you can specify a maximum of 16 CNTLUNIT statements
to specify the same pchid or the same link address for a pchid on an ESCON Director.

v For FC channel paths, you can specify a maximum of 256 CNTLUNIT statements to specify the
same pchid or the same link address for a pchid on a FICON Director.
v For an IQD channel path, you can specify a maximum of 64 CNTLUNIT statements.
v For an OSD, OSM, or OSX channel path, you can specify a maximum of 16 CNTLUNIT statements.
The OSAD device with unit address FE requires a control unit definition with a CUADD keyword
value of zero or no CUADD keyword. Otherwise, the device is unusable.
Note: A CNC or CTC channel path not assigned to an ESCON Director is viewed as having a single
link address.
The CUADD keyword is valid only for: CNC, CTC, FCV, FC, IQD, OSD, OSM, and OSX channel
paths (TYPE keyword).
For a channel path assigned to an ESCON CTC or FICON CTC control unit, the logical address
identifies the logical partition to which this channel path is to communicate. If the remote channel
path to which this channel path must communicate is not shared, then the logical address must be
zero or not specified. If the remote channel path to which this channel path must communicate is
shared, then the logical address must equal the wanted MIF image ID1 from the RESOURCE
statement for the logical partition that the shared channel path can access. This applies to CNC, CTC,
FCV, and FC remote channel paths that are shared.
However, if the remote channel path is a shared FC channel path on a CPC that supports multiple
logical channel subsystems (that is, if the IOCP described in this document supports the CPC), then
the logical address must equal the combination of the wanted CSS ID and the MIF image ID for the
logical partition with which you want to communicate. For example, if the remote channel path is
shared within CSS 1 and you want to communicate with the logical partition that has MIF image ID
5, specify CUADD=15.
ESCON CTC and FICON CTC communications are on a logical partition basis even if a channel path
is shared by multiple logical partitions. You must define a control unit for every logical partition to
which you want to communicate.
Note: The preceding CTC requirements for the CUADD keyword apply to any type of remote CPC
(for example, a 9672 CPC).
For more information about ESCON CTC and FICON CTC communication, see ESCON and FICON
Channel-to-Channel Reference, SB10-7034.
In addition to the preceding syntax rules, the following configuration rules also apply when
specifying the CUADD keyword:
v All CUADD values must be unique for a given path (pchid.link) regardless of whether a director
was specified. That is, the combination pchid.link.cuadd is allowed only once in the configuration.
See“Example 6” on page 80 for CNTLUNIT definition considerations.
v For an IQD, OSD, OSM, or OSX channel path, you cannot duplicate CUADD values. Every control
unit on the channel path must specify a unique CUADD value.
v If the CUADD keyword is specified, the CUADD keyword must also be specified on all other
CNTLUNIT statements that have at least one path in common (same switch.link or pchid.link).
v If a channel path does not attach to a director (SWITCH keyword), specifying FICON CTC and
non-FICON CTC control units on the same channel path is not allowed. See “IOCP definitions” on
page 56. The control units must either all specify UNIT=FCTC or none of them can specify FCTC.
v All CUADD values must be unique for a given path through a director (switch.link) unless the
following conditions are true:
1. For IOCP programs not described by this document, the MIF image ID was called a partition number.

– The control units are ESCON CTC or FICON CTC control units (see “IOCP definitions” on page
56), and the channel paths for each control unit are all the same type or a mixture of CNC and
FCV.
– The control units have the same unit address range. The channel paths in each control unit are
all the same type or a mix of CNC and FCV, and none of the paths attaching to the different
control units belongs to the same logical partition. The logical partition intersection test is based
on the initial access lists of the channel paths attaching to the control units.
Otherwise, the combination switch.link.cuadd is allowed only once in the configuration.
PROTOCL=
Specifies the interface protocol that the parallel control unit uses when operating with the channel
paths specified in the PATH parameter. The PROTOCL keyword is meaningful only for CVC channel
paths (TYPE keyword). IOCP ignores the PROTOCL keyword for all other channel path types. IOCP
assigns PROTOCL=D to control units attached to CBY channel paths.
Note: Incorrect specification of the PROTOCL parameter might produce unpredictable results. For
example, if you specify PROTOCL=S (data streaming) for a control unit that supports the DC
interlock protocol, I/O requests can result in detected errors (such as data overruns or interface
control checks) or in undetected errors. Additionally, specifying PROTOCL=S for a control unit that
runs at a maximum data rate of 4.5 MB per second could cause bus parity checks.
D Specifies the direct-coupled interlock (DC interlock) protocol. This protocol is the standard I/O
interface that requires a response to a demand. If you do not code PROTOCL, D is the default.
S Specifies data streaming protocol at a maximum data rate of 3.0 mps (megabytes per second).
This protocol does not require the demand response but continues at the rate governed by the
control unit.
S4 Specifies data streaming protocol at a maximum data rate of 4.5 mps. This protocol does not
require the demand response but continues at the rate governed by the control unit.
SHARED=
Specifies the level of concurrency of I/O requests that the parallel channel path allows for the control
unit. IOCP automatically sets the control unit type (1 or 2) based on the SHARED parameter that you
specify. See Appendix D, “List of I/O devices and control units,” on page 245 for more information
about specifying the SHARED parameter. (For further information about control unit types, see
“Control unit types” on page 45.)
The SHARED keyword is meaningful only for CVC channel paths (TYPE keyword). IOCP ignores the
SHARED keyword for all other channel path types. Control units that attach to CBY, CNC, CTC, FCV,
FC, FCP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel paths are type 2. Control units that
attach to CFS, CFP, CBS, CBP, CIB, CL5, CS5, and ICP channel paths are type 1.
Y Specifies that the control unit supports:
v Only one I/O request at a time, regardless of the number of I/O devices attached (for example,
the 3803 Tape Control Unit for magnetic tape units or the 3258, 3272, or 3274 Control Unit)
v Disconnect command chaining (DCC) operations.
IOCP assigns the control unit as type 1.
The channel ensures that only one I/O request is processed at a time for all devices sharing the
control unit. If sense information is pending in the control unit for an I/O device attached to the
control unit, the channel ensures that I/O requests for other I/O devices attached to the control
unit are not initiated until the system control program has issued an I/O request to clear the
pending sense information.
N Specifies that the control unit supports multiple I/O requests concurrently (one for each attached
I/O device). IOCP assigns the control unit as type 2. Specifying SHARED=N for a control unit
that does not support multiple concurrent requests can result in a control unit malfunction.

UNIT=type
Specifies the type of control unit. You can specify as many as eight alphanumeric or special (-, #, @,
or $) characters that identify the type of control unit. IOCP checks the syntax for alphanumeric or
special characters.
IOCP takes special actions for control unit types SCTC and FCTC. See the LINK and CUADD
keyword sections for the CNTLUNIT statement. Also, control unit type FCTC is valid only for FC
channel paths (TYPE keyword).
You must specify a type for every control unit in your I/O configuration.
A Server Time Protocol (UNIT=STP) control unit with CIB, CL5, or CS5 channel paths (TYPE
keyword) must have only external connections. No CIB, CL5, or CS5 channel path in an STP control
unit can have a CSYSTEM keyword value that matches the local system name (LSYSTEM keyword)
of the CPC, forming an internal connection. Also, an STP control unit cannot have ICP channel paths.
Notes:
1. To define a FICON channel-to-channel (CTC) connection between FC channel paths (TYPE
keyword), specify UNIT=FCTC for each CTC control unit defined on the FC channel path.
2. For integrated control unit/devices (such as the IBM 3800 Page Printer), specify the same type
value in the UNIT parameter of the CNTLUNIT statement as in the UNIT parameter of the
IODEVICE statement. For example, for the IBM 3800 Page Printer, specify UNIT=3800 in each of
the two statements.
UNITADD=
address
Specifies the unit addresses of the I/O devices that the control unit recognizes. You specify two
hexadecimal digits in the range of 00-FF. For example, UNITADD=((0A),(0B),(0F)) specifies that
the control unit recognizes unit addresses 0A, 0B, and 0F. At least one unit address must be
specified. A unit address does not need to represent an attached device.
number
Specifies the number of sequential unit addresses the control unit recognizes. You specify a 1-, 2-,
or 3-digit decimal value in the range of 1-256. For example, UNITADD=((0A,3)) indicates that the
control unit recognizes unit addresses 0A, 0B, and 0C. The unit address plus the number of
addresses must not exceed an address of FF. If the number is omitted, a value of 1 is assumed.
You can specify a maximum of eight addresses or sets of address and number values for parallel
control units. For ESCON or FICON control units, you can specify only one range of unit addresses.
The unit addresses specified for the CNTLUNIT statement must include the unit addresses of the
attached devices. The unit address of a device is taken from the UNITADD parameter of the
IODEVICE statement if specified or from the rightmost two digits of the device number specified
with the ADDRESS parameter of the IODEVICE statement.
Note: See Appendix D, “List of I/O devices and control units,” on page 245 for some exceptions and
recommendations that apply to specific device types. Some ESCON and FICON control units require
unit address range to begin with (hexadecimal) 00.
IOCP ignores the UNITADD keyword for control units attached to CFS, CFP, CBS, CBP, CIB, CL5,
CS5, FCP, ICP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel paths. IOCP defines unit addresses
00-FE for control units attached to OSD, OSE, OSM, OSN, or OSX channel paths. IOCP defines unit
addresses 00-FD for control units attached to FCP or OSC channel paths. IOCP defines unit addresses
00-FF for control units attached to IQD channel paths.
The full range of unit addresses that the control unit can address must be specified regardless of
whether devices are attached. If you do not specify one or more of the unit addresses that the control
unit can address, missing interruption handler (MIH) situations result. However, this requirement
does not apply to ESCON CTC or FICON CTC control units.

Notes:
1. Specify identical unit addresses in the CNTLUNIT statements for TYPE=CTC and TYPE=CNC or
TYPE=FCV channel paths to communicate with each other. Specify identical unit addresses for
FICON CTC control units to communicate with each other.
2. Unit addresses must be unique for each chpid on parallel control units. Two control units attached
to the same parallel channel path must have unique unit address ranges.
3. Unit addresses must be unique for each control unit (not for each chpid) for ESCON, FICON, IQD,
OSD, OSM, and OSX devices. Two devices attached to the same ESCON, FICON, FC, IQD, OSD,
OSM, or OSX control unit cannot have the same unit address. However, two devices attached to
the same CNC, CTC, FC, FCV, IQD, OSD, OSM, or OSX channel path (TYPE keyword) can have
the same unit address if they attach to different control units.
A shared CTC channel path (SHARED keyword specified in the CHPID statement) can generate more
unit addresses and unit address ranges than an unshared channel path. See “Shared ESCON CTC
A spanned FC channel path (PATH keyword in the CHPID statement) can generate more unit
addresses and unit address ranges than an unspanned channel path. See “FC control unit definition
considerations” on page “FC control unit definition considerations:” on page 86.
CNTLUNIT statement coding examples
Example 1
CU10A CNTLUNIT CUNUMBR=10A,PATH=06,N, X
UNIT=3800,UNITADD=((0A,3)),PROTOCL=S

v Assigns the control unit number of 10A to an IBM 3800 Page Printer.
v Names channel path 06 to which the printing subsystem is attached. The CSS parameter was not
specified for the PATH keyword because there is only one CSS in the configuration.
v Specifies the 3.0 MB data-streaming protocol.
v Defines unit addresses 0A-0C to the control unit.
Example 2
CU180 CNTLUNIT CUNUMBR=180,PATH=05,PROTOCL=S4, X
N,UNIT=3490,UNITADD=((80,16))

v Assigns the control unit number of 180 to an IBM 3490 Magnetic Tape Subsystem.
v Names channel path 05 to which the control unit is attached. The CSS parameter was not specified for
the PATH keyword because there is only one CSS in the configuration.
v Specifies the 4.5 MB data-streaming protocol.
v Defines unit addresses 80-8F to the control unit.
Example 3
CU11A CNTLUNIT CUNUMBR=11A,PATH=(06,07),LINK=(E0,E1) X
UNIT=3990,CUADD=1,UNITADD=((00,16))

v Assigns the control unit number of 11A to an IBM 3990 Storage Control.

v Names two channel paths (06 and 07) to which the control unit is attached. (These paths are defined as
TYPE=CNC and are assigned to an ESCON Director in the CHPID statement associated with this
CNTLUNIT statement.) The CSS parameter was not specified for the PATH keyword because there is
only one CSS in the configuration.
v Assigns link address E0 to channel path 06 and link address E1 to channel path 07.
v Defines logical address 1.
Example 4
CU987 CNTLUNIT CUNUMBR=987,PATH=3F,LINK=F0, X
UNIT=SCTC,UNITADD=((40,16))

v Assigns 987 as the control unit number for an ESCON CTC control unit.
v Assigns the control unit to channel path 3F in the only CSS in the configuration. (The path was defined
as TYPE=CTC and assigned to an ESCON Director in the CHPID statement associated with this
CNTLUNIT statement.)
v Assigns link address F0 to the channel path.
Example 5
CNTLUNIT CUNUMBR=290,PATH=((CSS(0),34,79),(CSS(1),80)), X
LINK=((CSS(0),FE,FE),(CSS(1),FE)),UNIT=SWCH, X
UNITADD=00

v Assigns control unit number 290 to an ESCON Director Model 1.
v Assigns the control unit to channel paths 34 and 79 in CSS 0 and channel path 80 in CSS 1. (These
paths are defined as TYPE=CNC in the CHPID statement associated with this CNTLUNIT statement.)
v Assigns link address FE to all three channel paths.
v Defines unit address 00 to the control unit.
Example 6
CHPID PCHID=100,PATH=(CSS(0,1),40),TYPE=FC,SWITCH=25
CNTLUNIT CUNUMBR=280,UNIT=2105,UNITADD=((00,256)),CUADD=01, X
PATH=((CSS(0),40),(CSS(1),40)), X
LINK=((CSS(0),D0),(CSS(1),D0))
The preceding statement performs the following actions:

v Define a spanned FC channel path.
v Assigns 280 as the control unit number of the IBM 2105 Enterprise Storage Server.
v Defines unit addresses 00-FF for the logical control unit.
v Defines logical address 01 for the logical control unit.
v Assigns the control unit to channel path 40 in CSSs 0 and 1.
v Assigns the same link address D0 to channel path 40 in both CSSs.
When configuring a spanned channel path to use the same link address in its CSSs for the same control
unit logical address (CUADD), define a single control unit containing all the CSSs for the channel path.
Do not define separate control units for each CSS for the channel path. Doing so would violate the rule

that the path combination pchid.link.cuadd can occur only once in the configuration. For example, control
unit 280 in the preceding example could not be defined as follows:
PATH=((CSS(0),40)),LINK=((CSS(0),D0))
PATH=((CSS(1),40)),LINK=((CSS(1),D0))
See Appendix E, “Configuration examples,” on page 253 for more CNTLUNIT statement coding
examples.
IODEVICE
IODEVICE is an optional statement that describes the following:
v The I/O device number
v The device characteristics
v The control units to which the device is assigned.
Each uniquely addressable I/O device in your I/O configuration must be specified in an IODEVICE
statement to use for I/O requests.
See Appendix D, “List of I/O devices and control units,” on page 245 for a list of some I/O device types
and their characteristics.
For z/VM: VM does not necessarily support devices that IOCP supports. For a list of supported devices,
see the General Information Manual for the VM operating system that you use
For z/VSE: z/VSE does not necessarily support devices that IOCP supports. For a list of supported
devices, see System Control Statements for the z/VSE operating system that you use.
See Appendix F, “Machine limits and rules,” on page 285 for a list of machine limits and rules for your
CPC.
The IODEVICE statement must follow any CNTLUNIT statements that specify the control units the
device uses.
The format of the IODEVICE statement is as follows:

[symbol] IODEVICE ADDRESS=(number[,count])
CUNUMBR=(number[,number]...)
[MODEL=model]
[{PART | PARTITION | NOTPART}=
((CSS(cssid),{name|0}[,name]...),...)]
[PATH=((CSS(cssid),chpid),...)]
[SCHSET={ssid|((CSS(cssid),ssid),...)}]
[STADET={Y|N}]
[TIMEOUT={Y|N}]
UNIT=device
[UNITADD=address]
ADDRESS=
number
Specifies the device number. Specify a 1-4 digit hexadecimal number in the range 0000-FFFF.
For z/OS: Although z/OS supports the definition and use of devices with 4-digit device numbers,
not all products, subsystems, or applications that run under z/OS can use these devices. See the
MVS/ESA SP 5.1 Conversion Notebook , GC28-1436, for more information.
For z/VSE: Prior to z/VSE 4.3, the supported address range is 0000-0FFF. For z/VSE 4.3 and later,
the supported address range is 0000-FFFF. Devices with 4-digit device numbers can be ADDed
during IPL. The z/VSE system continues to work with 3-digit device numbers. Therefore, during
IPL the z/VSE system assigns each 4-digit device number a 3-digit one. The 3-digit device
number that is to be used by z/VSE can also be explicitly specified by the user. Applications,
commands, messages continue to use the assigned 3-digit device number.
If you do not specify the UNITADD keyword in the IODEVICE statement, the last two
hexadecimal digits in the device number specify the unit address that is transmitted on the
channel path to select the device. These two hexadecimal digits must match one of the unit
addresses specified in the UNITADD parameter of the CNTLUNIT statement that defines the
control units to which the device is attached.
If you specify the UNITADD parameter in the IODEVICE statement, the last two digits in the
device number can be any value that follows the rules in Appendix D, “List of I/O devices and
control units,” on page 245 and makes the device number unique.
count
Specifies the number of sequential device numbers to be assigned to the devices. You specify a 1-,
2-, or 3-digit decimal value in the range of 1-256. (For example, ADDRESS=(10A,3) specifies that
the device numbers of 10A, 10B, and 10C are assigned.)
The unit address for a device defaults to the last two digits in the device number parameter of
the ADDRESS keyword unless the UNITADD keyword is specified. The unit address plus the
number of addresses specified in the count parameter of the ADDRESS keyword, minus 1, must
not be larger than a hexadecimal value of FF. The device number specified in the ADDRESS
keyword plus the number of devices specified in the count parameter of the ADDRESS keyword,
minus 1, must not be larger than a hexadecimal value of FFFF.
The specified device numbers are defined in the appropriate subchannel set (SCHSET keyword) in
each CSS to which the attached control units are assigned.
FCP, IQD, OSD, OSM, OSN, and OSX channel paths support a maximum number of valid
subchannels. Valid subchannels are devices that are defined to a logical partition and permitted for
use by the logical partition. Valid subchannels are calculated as follows:
Also, for reconfigurable FCP channel paths on CPCs without fixed HSA, each device defined in
IOCP represents a single valid subchannel. See Table 16 on page 285 to determine if your CPC has
fixed HSA.
v For reconfigurable FCP channel paths on CPCs with fixed HSA, each device defined in IOCP
represents multiple valid subchannels based on the total number of logical partitions (LPs) in its
CHPID candidate list.
v For shared channel paths, each device defined in IOCP represents multiple valid subchannels based
on the total number of LPs in its device candidate list.
When priority specification is enabled (CHPARM=00), a maximum of 480 OSA valid subchannels (160
TCP/IP stacks) can be defined on an OSD or OSX channel path. When priority specification is
disabled (when CHPARM=02 is specified on CHPID statement), a maximum of 1920 OSA valid
subchannels (640 TCP/IP stacks) can be defined on an OSD channel path on an OSA-Express2 or
later feature and on an OSM or OSX channel path on an OSA-Express3 or later feature. An OSA
subchannel is a device with a unit address in the range 00-FD. The OSAD device with unit address
FE is excluded from the maximum.
A maximum of 480 OSN valid subchannels and 180 3745 valid subchannels can be defined on an
OSN channel path. OSN subchannels are defined with UNIT=OSN on the IODEVICE statement, and
3745 subchannels are defined with UNIT=3745. Valid subchannels are devices with unit addresses in
the range 00-FD. The OSAD device with unit address FE is excluded from the maximums. A
maximum of 480 valid subchannels can be defined on an FCP channel path.

A maximum of 12288 valid subchannels can be defined on all IQD channel paths in the configuration.
The maximum applies to the valid subchannels for all IQD channel paths.
Multiple I/O devices with the same device number can be defined as long as device numbers are not
duplicated within the same logical partition and the same subchannel set. IOCP determines whether
devices in the same subchannel set with duplicate device numbers belong to the same logical
partition by testing the initial access lists of the chpids assigned to the devices along with the device
candidate lists (PARTITION or NOTPART keyword in the IODEVICE statement). A duplicate device
number error occurs if an initial access list of a chpid assigned to each device includes the same
logical partition and the device candidate lists of each device also include the same logical partition
and the devices are in the same subchannel set.
Note: See Appendix D, “List of I/O devices and control units,” on page 245 for some exceptions and
recommendations that apply to specific device types.
OSD, OSM, or OSX device definition considerations: For information about how to configure OSD,
OSM, or OSX devices, see “OSA configuration considerations” on page 33.
For VM: The ADDRESS keyword of the IODEVICE statement in the IOCP input file has values
derived from the RDEVICE statement in the SYSTEM CONFIG file, if used. For example, this
RDEVICE statement is in the SYSTEM CONFIG file:
RDEVICE 190-193 ...
The corresponding IODEVICE statement is:

IODEVICE ADDRESS=(190,4),...
For z/VSE: The ADDRESS keyword of the IODEVICE statement in the IOCP input file has values
derived from the ADD statement in the ASI Procedure file. For example, this ADD statement is in the
ASI Procedure file:
ADD 0C0:0DF,...

IODEVICE ADDRESS=(0C0,32),...
If you specify IODEVICE ADDRESS=(4900,32), ..., the corresponding ADD statement would be:
ADD 4900:491F as cuu1:cuu2
where ccu1, cuu2 is the range of 3-digit device numbers that are used by the z/VSE system.
CUNUMBR=number
Specifies the control unit numbers assigned to the control units (specified in the CUNUMBR
parameter of the CNTLUNIT statement) to which the device is attached. You can specify one to four
hexadecimal digits in the range 0000-FFFE for each control unit.
The device is defined in each CSS to which the specified control units are assigned. A device can be
assigned to a maximum of eight control units and eight paths in a CSS with the following exceptions:
v A device assigned to an ESCON CTC or FICON CTC control unit can be assigned to only one
control unit and one path in a CSS.
v A device assigned to CFS, CFP, CBS, CBP, ICP, CIB, CL5, and CS5 channel paths can be assigned to
only one control unit.
v A device assigned to an FCP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel path can be
assigned to only one control unit and one path in a CSS.
When the device is attached to more than one control unit (for example, through a string switch),
each control unit must be specified.
If a device is attached to more than one control unit, all control units that attach to the device must:
v Use the same unit address to access the device
v Have the same SHARED characteristic: Y or N

v Have the same PROTOCL characteristic: D, S, or S4
v Not be attached to the same channel path in a CSS
v Attach to all shared channel paths or all unshared channel paths.
All channel path types (TYPE keyword) assigned to a device must meet the following restrictions:
v All CBY or CVC
v CTC
v All CFS or CBS
v OSC
v OSD
v OSE
v OSM
v OSN
v OSX
v FCP
v IQD
You can intermix FC channel paths with CNC and FCV channel paths. Mixing FC channel paths with
CNC and FCV channel paths can ease migration from ESCON channels to FICON channels using
dynamic I/O configuration. The mixture allows you to dynamically add FC channel paths to a
control unit while keeping its devices operational. A second dynamic I/O configuration change can
then remove the CNC and FCV channels while keeping the devices operational.
For some types of control units and devices, there might be small operational performance differences
when intermixing (for example, RMF usage calculations and channel path reconnection), and
dynamic CHPID management cannot manage a control unit that contains intermixing. For these
control units and devices, mixing of FC channel paths with CNC and FCV channel paths be
temporary. Non-dynamic I/O users are advised to avoid mixing these channel paths. Other types of
control units and devices fully support this intermixing. To determine the level to which your control
units and devices support intermixing, see their device planning manuals. For control units that
support intermixing, consult their performance white papers for guidance on how to best configure
channels from a performance perspective.
To improve the balance of workload across channels in an intermixed environment, alternate ESCON
(CNC and FCV) and FICON (FC) channel paths in the control unit header for this device using the
following patterns:
v For 6 ESCON and 2 FICON, use a sequence of E, E, E, F, E, E, E, F
v For 4 ESCON and 4 FICON, use a sequence of E, F, E, F, E, F, E, F
v For 2 ESCON and 6 FICON, use a sequence of E, F, F, F, E, F, F, F.
For VM: To code the control unit number for a device, you must find the correct CNTLUNIT
statement. You must know the following information about the hardware I/O configuration:
v The channel path to which the device and its control unit are attached
v How the control unit recognizes the device, that is, by which unit address
Find the CNTLUNIT statement that fulfills both of the following requirements:
v A channel path that matches the channel path to which the device and control unit are attached
v A unit address that matches the unit address by which the control unit recognizes the device.
Use the value coded for the CUNUMBR parameter in the CNTLUNIT statement for the control unit
number in the IODEVICE statement.

For example, eight 3380 Direct Access Storage units and their control unit are attached to chpid 07.
The control unit recognizes the direct access storage units as attached at unit addresses 40-47.
However, the RDEVICE statement for this group of direct-access storage units could be:
RDEVICE 350-357 TYPE 3380
The device numbers for these disk storage units are in the range 350-357.
The CHPID statement coded for this channel path is:
CHPID PATH=(07),...
The chpid is 07. The physical unit addresses of these devices are 40-47. The CNTLUNIT statement that
has channel path 07 and unit addresses in the range 40-47 is:
CNTLUNIT CUNUMBR=032,PATH=(07),...,UNITADD=((40,8))
The CUNUMBR parameter in the IODEVICE statement is 032:
IODEVICE ADDRESS=(350,8),CUNUMBR=032,...,UNIT=3380,UNITADD=40
(The UNITADD parameter in the IODEVICE statement has the same number of unit addresses (eight)
as the number specified in the ADDRESS parameter.)
This example uses the UNITADD parameter, which allows you to define device numbers (350-357)
that do not contain their physical unit addresses (40-47).
MODEL=model
Specifies the model number, if any, for the device. You can specify a maximum of four alphanumeric
or special characters (#, @, or $) that represent the model number of the device. IOCP checks only the
syntax for alphanumeric or special characters. IOCP does not validate the model value.
For VM: The MODEL=model parameter of the IODEVICE statement in the IOCP input file has the
same value as the MODEL=model parameter of the RDEVICE statement in the SYSTEM CONFIG file.
For example, the following RDEVICE statement is in the SYSTEM CONFIG file:
RDEVICE ... MODEL 11
IODEVICE ...,MODEL=11,...
PART | PARTITION=
NOTPART=
Specifies the device candidate list that identifies the logical partitions (LPs) in a CSS that can access
the device. An LP cannot access a device if the LP is not specified in the device candidate list for the
device, even if the LP can access a channel path assigned to the device.
The PART or PARTITION keyword specifies the LPs that are in the device candidate list. The
NOTPART keyword specifies the LPs that are not in the device candidate list. For example, if a CSS
has three LPs (LP1, LP2, and LP3) and you specify NOTPART=(LP2), LP1 and LP3 can access the device,
but LP2 cannot.
cssid
your CPC supports.
Each CSS ID must be associated with a control unit attached to this device. You can specify as
many CSSs as the attached control units have. If all the attached control units are associated with
the same, single CSS ID, the CSS parameter is optional.
name
Specifies the name of a logical partition (LP) in a CSS that can access the channel path. The LP
name is a 1-8 alphanumeric (0-9, A-Z) character name that must have a nonnumeric first
character.

0 Specifies that no LPs in the CSS can access the device. A zero indicates a null device candidate
list. The device is excluded from the specified CSS and cannot be accessed by any LP in the CSS.
A zero cannot be specified in the NOTPART keyword.
OSC, OSD, OSE, OSM, OSN, and OSX device definition considerations: For information about
how to configure OSC, OSD, OSE, OSM, OSN, and OSX devices, see “OSA configuration
IQD device definition considerations: For information about how to configure IQD devices, see
“Internal queued direct communication (HiperSockets) support” on page 36.
FCP device definition considerations: For information about how to configure FCP devices, see
“Fibre Channel Protocol support” on page 37.
Device candidate list specification rules: The following rules apply when specifying the device
candidate list:
v You cannot specify a device candidate list using both the NOTPART keyword and either the PART
or the PARTITION keyword for the same CSS ID in the same IODEVICE statement.
v You can specify a device candidate list only for shared CNC, CTC, FCV, FC, FCP, IQD, OSC, OSD,
OSE, OSM, OSN, or OSX channel paths. You cannot specify a device candidate list for unshared
channel paths or CFS, CFP, CBS, CBP, ICP, CIB, CL5, and CS5 channel paths.
v Device candidate list specification is optional.
v If you do not specify a device candidate list for a CSS to which the device is assigned:
– Devices assigned to a shared CTC, FCP, IQD, OSD, OSM, OSN, or OSX channel path have a
default device candidate list equal to the candidate list for the channel path. This statement
excludes reserved LPs (*). The LPs in the CTC, FCP, IQD, OSD, OSM, OSN, or OSX channel path
candidate list are the only LPs in the default device candidate list.
– Devices assigned to all other shared channel paths types have a default device candidate list
equal to all LPs in each CSS with access to the device. This statement includes reserved LPs (*).
v If you specify a device candidate list for a CSS, it excludes any reserved LPs (*) in the CSS.
v You can specify all LP names in a CSS in the NOTPART keyword so that no LPs in the CSS can
access the device. Doing so creates is a null device candidate list and is equivalent to specifying a
zero in the PART or PARTITION keyword.
v You cannot specify null device candidate lists for every CSS to which the device has access. Doing
so is equivalent to not defining the device.
v You cannot specify null device candidate lists for every CSS to which one of the device's control
units have access. Doing so excludes the control unit from accessing the device. Instead, simply
omit the control unit from the CUNUMBR keyword for the device.
v You cannot specify null device candidate lists so that every device assigned to a control unit
excludes the same CSS. Channel paths in the excluded CSS of the control unit would not be able to
access the control unit or any of its devices. Instead, simply omit the CSS from the PATH keyword
for the control unit.
v If you specify a zero for a null device candidate list in a CSS, you cannot specify a logical partition
name for the same CSS.
v You cannot specify an LP in the device candidate list that is not defined in the RESOURCE
statement.
v You can specify as many LPs as your CPC supports. See Appendix F, “Machine limits and rules,”
on page 285.
v You can specify any LP in the configuration in the device candidate list as long as at least one of
the LPs specified matches an LP in the channel path candidate lists for the channel paths assigned
to the device. When building an IOCDS, IOCP sends a caution message identifying every LP in the
device candidate list that cannot access the device. After power-on reset, the only way for an LP
that has no access to the device to gain access is through a dynamic I/O configuration change that
does one of the following actions:

– Modifies the channel path candidate lists of the channel paths assigned to the device
– Modifies the device candidate list
– Adds a channel path to a control unit assigned to the device where the LP can access the
channel path.
For example, if a device is assigned to a single channel path and the channel path has access to
LPs LP1 and LP2, you cannot specify a device candidate list of LP3. However, you could specify
LP1 and LP3. In this case, IOCP sends a caution message that LP3 cannot access the device.
Using the device candidate list to limit logical path establishment: You can use the device
candidate list to limit the establishment of logical paths to a control unit by not allowing an LP to
access a device on the control unit. See “Managing the establishment of logical paths” on page 41 for
more information.
PATH=
cssid
your CPC supports.
Each CSS ID must be associated with a control unit attached to this device and must not have
been excluded by specifying a null device candidate list with the PART, PARTITION, or
NOTPART keyword. You can specify as many CSSs as the attached control units have. If all the
attached control units have the same, single CSS ID, the CSS parameter is optional.
chpid
Specifies preferred channel paths for the CSSs that can access the device. Specify two hexadecimal
digits for a chpid assigned to the device for each wanted CSS.
IOCP ignores the PATH keyword, if specified, for devices assigned to CFS, CFP, CBS, CBP, CIB,
CL5, CS5, CTC, FCP, ICP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel paths (TYPE
keyword).
If specified, a CSS always first tries to use the preferred channel path to initiate I/O requests for
the device. If the preferred path is busy or unavailable, the CSS uses the other attached channel
paths. See “Path selection using a preferred path” on page 44 for more information.
SCHSET=
Specifies the subchannel set that contains the device. A subchannel set other than set 0 is known as a
alternate subchannel set. z/OS supports defining parallel access volumes (PAVs) and mirroring
secondary devices in an alternate subchannel set. Linux on System z also supports devices in an
alternate subchannel set. IOCP does not restrict the types of devices that can be defined in an
alternate subchannel set except as described in the information that follows.
cssid
your CPC supports.
Each CSS ID must be associated with a control unit attached to this device. It must not have been
excluded by specifying a null device candidate list with the PART, PARTITION, or NOTPART
keyword. You can specify as many CSSs as the attached control units have. If the device is in the
same subchannel set in each CSS that can access it, the CSS parameter is optional (for example,
SCHSET=1).
ssid
Specifies the subchannel sets that contain the device. The SS ID is a 1-digit number in the range
0-n. See Table 18 on page 287 to determine the maximum SS ID your CPC supports. Some CPCs
support multiple subchannel sets in a CSS.

Subchannel sets within a CSS are separate from each other and support a maximum number of
devices. See Table 18 on page 287 to determine the maximum number of devices each subchannel
set on your CPC supports. A logical partition can have duplicate device numbers, provided they
are in different subchannel sets.
If you do not specify the SCHSET keyword, the default SS ID is 0. If you specify the CSS
parameter but do not specify every CSS that can access the device, the device is defined in
subchannel set 0 for the unspecified CSSs. Therefore, the SCHSET keyword is required only to
define a device in a subchannel set other than 0 in one or more CSSs.
Devices assigned to CBY, CVC, CFS, CFP, CBS, CBP, ICP, CIB, CL5, or CS5 channel paths can be
defined only in subchannel set 0.
Devices with different unit addresses attached to the same control unit can be defined in different
subchannel sets in the same CSS. A single unit address on a control unit might not be associated
with different devices in different subchannel sets in the same CSS.
STADET=
Allows disabling of the Status Verification Facility. STADET=Y enables the facility; STADET=N
disables the facility. The default for parallel devices is STADET=N. The default for ESCON and
FICON devices is STADET=Y.
IOCP ignores the STADET keyword, if specified, for devices assigned to CBS, CBP, CFS, CFP, CIB,
CL5, CS5, FCP, ICP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel paths.
The Status Verification Facility provides the system with a means of indicating that it has been
presented with device status that had a valid checking block code (CBC) but that contained a
combination of bits that was inappropriate when the status byte was presented to the I/O subsystem.
The indication is called device-status check.
If a system that provides the Status Verification Facility replaces a system that does not provide the
facility, the replacing system might generate device-status-check conditions for events that the
replaced system ran without error. In this situation, it might be necessary to disable the facility for
the affected subchannel.
If a device can generate busy and status-modifier status in response to initial selection during
command chaining, the Status Verification Facility treats the status as an error when STADET=Y. It
does not treat the status as an error when STADET=N.
TIMEOUT=
Specifies whether the I/O interface timeout function is active for the following I/O interface
sequences between the channel and the I/O device:
v 4-second timeout for initial selection
v 30-second timeout for data transfer.
The TIMEOUT keyword is meaningful only for CVC channel paths. IOCP ignores the TIMEOUT
keyword, if specified, for all other channel path types. IOCP assigns TIMEOUT=Y to devices assigned
to CBY channel paths.
Y Specifies making the I/O interface timeout function active for all sequences on the I/O interface
except as noted in Appendix C, “Characteristics of the I/O interface timeout function,” on page
243. With the timeout function active, if the I/O device fails to complete the initial selection
within 4 seconds, the channel ends the I/O request to the I/O device and generates an
interface-control-check interruption.
TIMEOUT=Y is the default if you do not code the TIMEOUT parameter, provided the keyword is
not ignored for the type of channel path assigned to the device.
N Specifies making the I/O interface timeout function inactive (timeout is inhibited) for some of the
sequences on the I/O interface as described in Appendix C, “Characteristics of the I/O interface
timeout function,” on page 243. With the timeout function inactive, the channel waits until the
I/O device completes the tag sequence or initiates the next dependent sequence.

UNIT=device
Specifies the device type. You specify a maximum of 8 alphanumeric or special characters (#, @, or $)
to define the device type (such as, 3390 or 3590). IOCP checks the syntax only for alphanumeric or
special characters. IOCP does not validate the device value except for devices assigned to IQD, OSD,
OSE, OSM, OSN, or OSX channel paths. For OSD, OSE, OSM, and OSX channel paths, ensure that
devices with unit addresses in the range 00-FD have a value of OSA and that the device with unit
address FE has a value of OSAD. For OSN channel paths, IOCP requires devices with unit addresses
in the range 00-FD to have a value of OSN or 3745, and IOCP expects the device with unit address
FE to have a value of OSAD. For IQD channel paths, ensure all devices have a value of IQD.
If you code UNIT=33xxx or 9345x and the device attaches to CVC channel paths (TYPE keyword), the
I/O subsystem activates unlimited prefetching of channel command words (CCWs). For more
information, see Enterprise Systems Architecture/390: Principles of Operation. To run applications that
require parallel basic mode CTC support in an ESCON environment, specify UNIT=BCTC. Any other
specification in an ESCON environment results in extended mode CTC support. For more
information about ESCON CTC support for parallel CTC basic mode, see ESCON and FICON
Channel-to-Channel Reference, SB10-7034.
See “OSA configuration considerations” on page 33 for more information concerning the
configuration of specific devices.
For VM: The UNIT=device parameter of the IODEVICE statement in the IOCP input file has the same
value associated with the TYPE parameter of the RDEVICE statement in the SYSTEM CONFIG file.
For example, assume that the following RDEVICE statement is in the SYSTEM CONFIG file:
RDEVICE 100 TYPE 3279

IODEVICE ...,UNIT=3279,...
UNITADD=address
Specifies the unit address that is transmitted on the channel path to select the I/O device. You specify
two hexadecimal digits in the range of 00-FF. (The control unit address must be physically set to
recognize the address that is on the channel path.) These two hexadecimal digits must match one of
the unit addresses specified in the UNITADD parameter of the CNTLUNIT statements that define the
control units to which the device is attached.
UNITADD allows you to assign a unit address to a device number in the ADDRESS parameter that
does not contain the actual physical unit address of the device. The UNITADD parameter that you
specify becomes the unit address that the control unit recognizes for the device. If you specified a
value for count in the ADDRESS parameter, that same number of sequential device numbers applies
to the UNITADD parameter. However, the UNITADD parameter plus the number of device numbers
minus 1, must not exceed a value of FF.
Coupling facility device definition considerations: IOCP ignores the UNITADD keyword, if
specified, for devices assigned to CFS, CFP, CBS, CBP, ICP, CIB, CL5, and CS5 channel paths.
OSC, OSD, OSE, OSM, OSN, and OSX device definition considerations: For information about
how to configure OSC, OSD, OSE, OSM, OSN, and OSX devices, see “OSA configuration
FCP device definition considerations: For information about configuring FCP devices, see “Fibre
Channel Protocol support” on page 37.
Notes:
1. If you use the UNITADD parameter, you must follow the same rules and restrictions that apply to
the last two digits of the ADDRESS parameter. For example, notes in Appendix D, “List of I/O
devices and control units,” on page 245 that apply to the unit address specified in the ADDRESS
keyword also apply to the unit address specified in the UNITADD keyword.

2. On parallel control units, unit addresses must be unique for each chpid. Two control units attached
to the same parallel channel path must have unique unit address ranges.
3. For ESCON or FICON devices, unit addresses must be unique for each control unit (not for each
chpid). Two devices attached to the same ESCON or FICON control unit cannot have the same
unit address. However, two devices attached to the same CNC, CTC, FCV, or FC channel path
(TYPE keyword) can have the same unit address if they attach to different control units.
4. Unit addresses must be unique for devices defined to an OSC, OSE, OSN, or FCP control unit.
Two devices defined to the same OSC, OSE, OSN, or FCP control unit cannot have the same unit
address.
5. Unit addresses must be unique for each control unit (not for each chpid) for IQD, OSD, OSM, or
OSX devices. Two devices defined to the same IQD, OSD, OSM, or OSX control unit cannot have
the same unit address. However, two devices defined to the same IQD, OSD, OSM, or OSX
channel path (TYPE keyword) can have the same unit address if they are defined to different
control units.
For VM: To code the UNITADD parameter, you must know how the control unit recognizes the
device, that is, by what unit address. You must code the actual physical unit address for the
UNITADD parameter in the IODEVICE statement.
For example, the RDEVICE statement for a group of eight 3380 Magnetic Disk Storage units could be:
RDEVICE 350-357 TYPE 3380
The device numbers for these disk storage units are in the range: 350-357. The control unit, however,
recognizes the disk storage units as being attached at unit addresses 40-47.
The UNITADD parameter in the IODEVICE statement is 40:
IODEVICE ADDRESS=(350,8),....UNIT=3380,UNITADD=40
The UNITADD parameter in the IODEVICE statement takes the same number (8) as specified in the
ADDRESS parameter. This example uses the UNITADD parameter to define device numbers (350-357)
that do not contain their physical unit addresses (40-47).
IODEVICE statement coding examples

Example 1
DEV00E IODEVICE ADDRESS=00E,CUNUMBR=10A,MODEL=3, X
UNIT=3800,TIMEOUT=Y

v Defines an IBM 3800 Page Printer Model 3 with an address of 00E.
v Assigns it to control unit number 10A.
Example 2
DEV2DD IODEVICE ADDRESS=(210,8),CUNUMBR=(1A0,1A1), X
UNIT=3390,TIMEOUT=Y

v Defines eight IBM 3390 Direct Access Storage devices.
v Assigns the addresses 210-217 to the devices.
v Assigns the devices to the control unit numbers 1A0 and 1A1, which represent two storage directors of
an IBM 3990 Storage Control.
Example 3
DEVDTT IODEVICE ADDRESS=(180,8),CUNUMBR=1EE, X
UNIT=3490,STADET=N,UNITADD=00

v Defines eight IBM 3490 Magnetic Tape Subsystems.
v Assigns the addresses 180-187 to the devices.
v Assigns the devices to control unit number 1EE, which attaches to ESCON channel paths (TYPE=CNC).
v Assigns unit addresses beginning at 00 to the devices for the control unit to recognize.
The default unit addresses for the devices are 80-87, and the control unit recognizes only unit addresses
00-0F. Specify the UNITADD keyword to make the device unit addresses 00-07.
Example 4
DEVDIR IODEVICE ADDRESS=(2C0),CUNUMBR=(290), X
UNIT=SWCH,UNITADD=00

v Defines an ESCON Director Model 2.
v Assigns the address 2C0 to the device.
v Assigns the device to control unit number 290.
v Assigns the unit address 00 to the device for the control unit to recognize.
The default unit address for the device is C0, and the control unit recognizes only unit address 00.
Specify the UNITADD keyword to make the device unit address 00.
Example 5
IODEVICE ADDRESS=(800,16),CUNUMBR=(800,801),UNIT=3390, X
PART=((CSS(0),LP1,LP2),(CSS(1),LPA))

v Defines 16 IBM 3390 Direct Access Storage devices.
v Assigns the addresses 800-80F to the devices.
v Assigns the devices to the control unit numbers 800 and 801, which represent an IBM 3990 Storage
Control's two storage directors and which have channel paths in CSSs 0 and 1.
v Specifies a device candidate list of LP1 and LP2 in CSS 0 and LPA in CSS 1 so that only those logical
partitions can access the devices.
Note: See Appendix E, “Configuration examples,” on page 253 for more IODEVICE statement coding
examples.
UUID
UUID is an optional statement that defines which logical partitions have UUID checking enabled. Logical
partitions with UUID checking enabled must have a unique user-defined identifier (UID) for each
function defined in your configuration with access to that logical partition. A UID can be specified for a
function using the UID keyword on the FUNCTION statement. If a function is assigned to a partition
with UUID checking enabled and does not have a UID defined, then IOCP assigns an arbitrary unique
UID value.
When specified, the UUID statement must

v Follow the ID statement (if specified) and the RESOURCE statement
v Precede all FUNCTION statements.
See Appendix A, “Coding IOCP statements,” on page 237 for a summary of rules for coding statements
and the notation in this document to illustrate statements.
The format of the UUID statement is as follows:

[symbol] UUID {PART | PARTITION}=({name[,name]...] | *})
PART | PARTITION=
Specifies the logical partitions for which UUID (unique user-defined identifier) checking is enabled
for functions. All logical partition (LP) names that you specify must match the ones specified in the
RESOURCE statement.
Each LP name is a 1-8 alphanumeric (0-9, A-Z) character name that must have a non-numeric first
character. If you want every LP in your configuration to have UUID checking enabled, you can
specify a single asterisk (*) instead of listing every LP name.
The following rules apply when UUID checking is enabled for a LP:
v Every function with a candidate list that includes that LP must have a unique UID (user-defined
identifier).
v You must either specify the UID keyword on the FUNCTION statement or let IOCP pick a default
UID value. The default UID values chosen by IOCP will not interfere with any specified UID
numbers.
v All UIDs in the LP must be unique.
v The UID number must be a maximum of 4 hexadecimal digits (for example, FFFF)
v When a function has a candidate list that contains both LPs that have UUID checking enabled and
LPs with UUID checking disabled, the uniqueness only applies to the partitions that have UUID
checking enabled. However, the UID value must be a maximum of 4 digits.
For LPs that do not have UUID checking enabled, a UID value is not required for every function.
Also, UID values that are specified can be duplicated for that LP.
UUID statement coding examples

Example 1
UUID PART=(LP1,LP3,LP4)
The preceding statement specifies that logical partitions LP1, LP3, and LP4 all have UUID checking
enabled. Any other logical partitions in the configuration do not have UUID checking enabled.
Example 2
UUID PART=*
The preceding statement specifies that every logical partition in the configuration has UUID checking
enabled.
FUNCTION
FUNCTION is an optional statement that defines:
v Function Identifier (FID)
v Physical Channel Identifier (pchid)
v Virtual Channel Identifier (vchid)
v Port on pchid
v Availability of FIDs to logical partitions
v Type of function
v Virtual Function number
v Physical network identifier
v User-defined identifier (UID).
When specified, the FUNCTION statement must

v Follow the ID and UUID statements (if specified) and the RESOURCE statement
v Collectivity, specify all functions to use in the configuration.
Functions not specified in a FUNCTION statement cannot be used.

A FUNCTION definition is a configuration wide definition. It is not associated with any one or groups of
logical channel subsystems (CSSs).
statements and the notation in this document to illustrate statements.
The format of the FUNCTION statement is as follows:

[symbol] FUNCTION FID=function number
[PCHID=number]
[VCHID=number]
[PORT=number]
PART=((access LP name),(candidate list))
[TYPE=function type]
[VF=virtual function number]
[PNETID=([,...]physical network id[,...])]
[UID=number]
FID=function number
Specifies a function identification (FID) number associated with an adapter. An adapter can have
multiple FIDs associated with it. The following rules apply to the function number specified.
v A 1- to 4- digit hexadecimal number in the range 0-n where n is the maximum number your CPC
supports. See Appendix F, “Machine limits and rules,” on page 285 to determine the maximum FID
number your CPC supports.
v The FID number is arbitrary and picked by the user.
v Each FID must be unique in the entire configuration.
PCHID=pchid number
Specifies the physical channel identification (pchid) number associated with the function. The pchid
number is a 1- to 3- digit hexadecimal number in the range 0-n where n is the maximum number
your CPC supports. See Appendix F, “Machine limits and rules,” on page 285 to determine the
maximum pchid number your CPC supports. The following rules apply to the pchid value specified:
v The pchid number identifies the physical location (cage and slot) for a function adapter in the CPC.
The CPC maintains this mapping.
v A PCHID on a FUNCTION statement must not match a PCHID on a CHPID statement or the
reverse.
For a given FUNCTION statement, either the PCHID or VCHID keyword must be specified. PCHID
and VCHID keywords are mutually exclusive and only one can be used on a given FUNCTION
statement. A PCHID is required for FUNCTION TYPEs of: HYL, RCE, ROCE, ROC2, and ZEDC.
The PCHID slots in I/O drawers for functions are assigned to resource groups. Some CPCs restrict
the number of ROCE and ROC2 PCHIDs (TYPE keyword) that should exist in a resource group.
Exceeding this limit can result in FIDs on those PCHIDs being unusable. To determine if your CPC
restricts the number of ROCE and ROC2 PCHIDs in a resource group and what the maximum is, see
Table 18 on page 287. To determine which PCHIDs and I/O slots are in the same resource group for
your CPC, see Table 3 on page 3.
VCHID=vchid number
Specifies the virtual channel identification (vchid) number associated with the function. The vchid
number identifies the (virtual) function in the CPC. The CPC maintains this mapping.
v The vchid number is a 1- to 3-digit hexadecimal number in the range n-m where n is the minimum
number and m is the maximum your CPC supports. See “Machine limits” on page 286 to
determine the minimum and maximum vchid numbers your CPC supports. If no vchid numbers are
listed, then the VCHID keyword is not supported on that CPC.
v A VCHID on a FUNCTION statement must not match a VCHID on a CHPID statement or the
reverse.
For a given FUNCTION statement, either the VCHID or PCHID keyword must be specified. VCHID
and PCHID keywords are mutually exclusive and only one can be used on a given FUNCTION
statement. VCHID is required for FUNCTION TYPE of ISM.

PORT=number
Specifies the port on the adapter to which FID is defined. The port number is a 1-digit decimal
number in the range 1-2.
The PORT keyword is valid only for HYL and ROC2 FIDs (TYPE keyword) and is required for both
of these types.
PART=
Specifies the availability of FIDs to logical partitions. All logical partition names that you specify
must match the ones specified in the RESOURCE statement.
access LP name
Specifies the logical partition (LP) that has the channel path configured online at LP activation
following initial power-on reset (POR) of an IOCDS. After the initial POR of an IOCDS, the I/O
subsystem retains which LPs have the FID configured online at LP activation following
subsequent PORs of the IOCDS. In this case, the access LP name used to build an IOCDS might
not match the access LP name the I/O subsystem retains. Logical partitions that are not specified
as the access LP name for a function can access the function after POR through reconfiguration
commands if the LP is specified in the candidate list for the function. (See “candidate list”).
{name| 0}
name
Specifies the name of a logical partition (LP) that can access the function. The LP name is a
1-8 alphanumeric (0-9, A-Z) character name that must have a nonnumeric first character.
0 Specifies that no LPs in the CPC can access the function following LP activation for the initial
POR of the IOCDS. The zero indicate null access.
v If a LP name is specified, it must have been previously defined on the RESOURCE statement.
v Only one LP name can be specified. Functions cannot be shared by multiple LPs at the same
time.
v If you specify a zero as the access LP name, you must specify a candidate list consisting of at
least one LP name.
candidate list
Specifies the logical partitions (LPs) that can access the function. LPs that are not specified in a
function candidate list of the path cannot access the function. If you specify a candidate list, you
do not need to repeat the LP name specified in the access LP name. IOCP automatically includes
in the candidate list the LP name specified in the access LP name.
{name [,name ]...|=}
name
See “name.”
= Specifies that the candidate list is identical to the access LP name.
The following rules apply when specifying the candidate list:

v You cannot specify an equal sign (=) for a candidate list of the function if the access LP name is
zero (0 indicates a null access).
v If you do not specify a candidate list, a candidate list defaults to all logical partitions defined in
the configuration, excluding reserved logical partitions (*).

VF=virtual function number
Specifies a virtual function number for a function adapter so that multiple logical partitions can use
the same adapter. Virtual function numbers are arbitrary. The following rules apply to the virtual
function number.
v The virtual function number is 1- to 3-digit decimal number in the range 1-n. See Table 10 for the
valid ranges for each function type.
v Function type RCE does not support virtual functions. Therefore, an RCE adapter can be used by
only one logical partition.
v For the 2827 and 2828 machine types, if a virtual function number is specified for a PCHID, then
| all definitions for the same PCHID must have a virtual function number specified. Also, these
| machine types do not support the TYPE keyword and the maximum VF number is 15. Therefore,
| Table 10 does not apply to these machine types.
v Beginning with the 2964 machine type, the VF keyword is required.
v A virtual function number cannot be duplicated for the same PCHID or VCHID or for the same
PCHID-PORT.
v HYL and ROC2 adapters have multiple ports and the VF values are on a port basis rather than an
adapter basis. All other function adapters have VF values on an adapter basis.
v Multiple virtual function numbers for the same PCHID or VCHID can be assigned to the same
logical partition.
The following virtual function ranges are supported for each function type:
Table 10. Valid VF ranges
Type Range
HYL 1-127 per port
ISM 1-255
RCE
ROCE 1-31
| ROC2 1-63 or 1-31 per port depending on CPC type (see 1)
ZEDC 1-15
1. See Table 18 on page 287 to determine the maximum VF number a ROC2 port on your CPC supports.
TYPE=function type
Specifies the type of function adapter. See Table 24 on page 295 to determine the type of functions
your CPC supports.
The TYPE keyword is required beginning with machine type (2965, 2964) and is ignored for prior
machines. The TYPE keyword value specified for multiple FUNCTION statements for the same
PCHID or VCHID must be identical. For function types that require the PORT keyword, ports on the
same PCHID must have identical function types. Optionally, once the TYPE keyword has been
specified for a PCHID or VCHID, subsequent FUNCTION statements can omit the TYPE keyword
and IOCP defaults to the type that was specified on the first FUNCTION statement for that PCHID
or VCHID.
The following values are the allowed Function types.
HYL
Specifies that the Function ID is associated with an IBM zHyperLinkTM Express adapter.
ISM
Specifies that the Function ID is associated with a SMC-D virtual network connection (internal).
RCE
Specifies that the Function ID is associated with a Regional Crypto Enablement (adapter) card. An
RCE adapter can have a single FID and is used by only one logical partition

ROCE|ROC
Specifies that the Function ID is associated with a 10GB RoCE (RDMA over Converged Ethernet)
adapter.
ROC2
Specifies that the Function ID is associated with a RoCE Express2 (RDMA over Converged
Ethernet) adapter.
ZEDC|EDC
Specifies that the Function ID is associated with a zEnterprise Data Compression adapter.
PNETID=physical network id
Specifies the physical network identifiers (pnetids) associated with the ports on the function adapter.
Each pnetid is a 1-16 alphanumeric (0-9, A-Z) value.
The pnetid in the PNETID keyword is a positional parameter. One to four pnetids can be specified
which correspond to the ports on the function adapter. You can omit one or more pnetids by
specifying just a comma for that adapter port position in the keyword value. For example, to only
specify the second pnetid for a function to be associated with the second port of the adapter, you
would specify PNETID=(,pnetid ).
Physical network IDs are used for the function types ROCE, ROC2, and ISM to establish the affinity
to a particular network. When network connections require two interfaces to be coordinated, as with
the 10GB RoCE network function adapter and an associated OSD channel path, the pnetids should
match.
The ROCE adapters have two ports so only one to two pnetids are allowed. An ISM function only
uses a pnetid associated with the first port.
The ROC2 adapter has two ports, but each port is defined on separate FUNCTION statements using
the PORT keyword. Therefore, only a single pnetid is allowed and must be in the first parameter
position of the PNETID keyword.
The physical network ID of IEDN is reserved for intra-ensemble data network (IEDN) ports or
network connections, and if present, applies to all ports.
For the ISM and ROCE function types, when multiple FIDs are defined for the same VCHID or
PCHID, the same PNETID value must be specified for all FIDs on that VCHID or PCHID. Also, when
PNETID values are specified for different VCHIDs for an ISM function type, all the PNETIDs must be
unique.
For the ROC2 function type, when multiple FIDs are defined for the same port on a PCHID, the same
PNETID value must be specified for all FIDs on that PCHID-PORT. The ports on a ROC2 PCHID can
have different PNETID values or no PNETID value.
If a pnetid is specified for the first FID defined for a PCHID, PCHID-PORT, or VCHID then
subsequent FIDs for that same PCHID, PCHID-PORT, or VCHID do not need to specify the PNETID
keyword and they default to the original pnetid.
UID=number
Specifies the user-defined identifier (UID) associated with the function. The UID number is a 1- to
4-digit hexadecimal number that is greater than zero. The UID number allows you to associate a
number to a function in addition to the FID.
The UID keyword is optional. However, when a logical partition (LP) in the function's candidate list
has unique user-defined identifier (UUID) checking enabled, every function with a candidate list that
includes that LP must have a unique UID. UUID checking is enabled for an LP by specifying the LP
name on the UUID statement.
The following rules apply when UUID checking is enabled for a LP:
v Every function with a candidate list that includes that LP must have a unique UID. All UIDs in
that LP must be unique.

v You must either specify the UID keyword on the FUNCTION statement or let IOCP pick a default
UID value. The default UID numbers chosen by IOCP will not interfere with any specified UID
numbers.
v When a function has a candidate list that contains both LPs that have UUID checking enabled and
LPs with UUID checking disabled, the uniqueness only applies to the partitions that have UUID
checking enabled.
For LPs that do not have UUID checking enabled, a UID value is not required for every function.
Also, UID values that are specified can be duplicated for the LP.
FUNCTION statement coding examples

Example 1
FUNCTION FID=1F,PCHID=330,PARTITION=((LP2),(LP6))

v Defines function identifier 1F for the card port associated with physical channel ID 330
v Specifies that logical partition LP2 initially accesses the function and that LP6 is also authorized to
access the function.
The above would be typical of a zEDC Express card on a machine type 2827 or 2828, which is dedicated
to one partition at a time, but is reconfigurable.
Example 2
FUNCTION FID=81,PCHID=360,PART=((LP6),(LP2)),VF=1
FUNCTION FID=82,PCHID=360,PART=((LP6),(LP3)),VF=2
FUNCTION FID=83,PCHID=360,PART=((LP3),(LP4,LP5)),VF=3
FUNCTION FID=8A,PCHID=360,PART=((LP6),(LP6,LP2)),VF=10
FUNCTION FID=8B,PCHID=360,PART=((LP6),(LP6,LP2)),VF=11
FUNCTION FID=8C,PCHID=360,PART=((LP6),(LP6,LP2)),VF=12
FUNCTION FID=8D,PCHID=360,PART=((LP6),(LP6,LP2)),VF=13
FUNCTION FID=8E,PCHID=360,PART=((LP6),(LP6,LP2)),VF=14
FUNCTION FID=8F,PCHID=360,PART=((LP8),(LP2,LP3,LP4,LP5,LP6)),VF=15

v Defines function identifier 81 through 8F for the card associated with physical channel ID 360.
v Specifies logical partition LP6 can initially access functions 81 - 84 and that logical partition LP2 is also
authorized to access the functions.
v Specifies logical partition LP6 can initially access functions 85 - 88 and that logical partition LP3 is also
authorized to access the functions.
v Specifies logical partition LP3 can initially access functions 89 - 8C and that logical partitions LP4 and
LP5 are also authorized to access the functions.
v Specifies logical partition LP8 can initially access functions 8D - 8F and that logical partitions LP2, LP3,
LP4, LP5, and LP6 are also authorized to access the functions.
The above would be typical of a zEDC Express card on a machine type 2827 or 2828 or later, which can
shared.
Example 3
FUNCTION FID=10,PCHID=314,PART=((Z13),(Z17,Z19,ZID)),TYPE=ROCE,PNETID=NETWORK1,VF=1
FUNCTION FID=11,PCHID=338,PART=((Z17),(Z13,Z19,ZID)),TYPE=ROCE,PNETID=(,NETWORK1),VF=5

CHPID PCHID=380,PATH=(CSS(0,1,2,3),D0),TYPE=OSD,SHARED,PNETID=NETWORK1
CHPID PCHID=3B8,PATH=(CSS(0,1,2,3),D4),TYPE=OSD,SHARED,PNETID=NETWORK1

v Defines function identifier 10 for a 10GB RoCE function adapter at physical channel ID 314. It is
initially used by logical partition (LP) Z13 and serially reconfigurable to LPs Z17, Z19, and ZID. Port 1
of the adapter is associated with physical network NETWORK1. Virtual function number 1 is assigned
to the FID.
v Defines function identifier 11 for a 10GB RoCE function adapter at physical channel ID 338. It is
initially used by LP Z17 and serially reconfigurable to LPs Z13, Z19, and ZID. Port 2 of the adapter is
associated with physical network NETWORK1. Virtual function number 5 is assigned to the FID.
v Defines CHPID D0 for physical channel ID 380 as an OSD channel path type. The first port of the
adapter is associated with physical network NETWORK1.
v Defines CHPID D4 for physical channel ID 3B8 as an OSD channel path type. The first port of the
adapter is associated with physical network NETWORK1.
v Are associated with a 2965 or 2964 or later CPC which requires the TYPE an VF keywords for ROCE
functions.
It is normal practice to pair up any 10GB RoCE features with an OSD channel path.
Example 4
FUNCTION FID=71,PCHID=33C,VF=1,TYPE=ROCE,PART=(LP6),(LP6,LP2))
The above statement:

Defines function identifier 71 for the card type ROCE associated with physical channel ID 33C.
Specifies that logical partition LP6 initially accesses the function, out of a partition candidate list of
LP2 and LP6.
The above would be typical of a ROCE card on a machine type 2965 or 2964 or above which can shared,
but in this instance is only shown for one partition.
IOCP functions
With IOCP, you can perform the following functions:
v Generate a new IOCDS from a file containing source IOCP statements
v Produce configuration reports from an existing IOCDS.
For z/OS: The rules for coding the JCL EXEC statements for generating an IOCDS in z/OS and for
producing configuration reports are in Chapter 3, “Running the z/OS version of IOCP,” on page 111.
For z/VM: The rules for using the IOCP command in VM to generate IOCDSs and to produce
configuration reports are in Chapter 4, “Running the VM version of IOCP,” on page 119.
For z/VSE: The rules for coding the JCL EXEC statements for generating an IOCDS in z/VSE and for
producing configuration reports are in Chapter 5, “Running the z/VSE version of IOCP,” on page 131.
Generating an IOCDS
IOCP performs the following steps when you specify that you want to generate an IOCDS.
Note: Steps 1-5 on page 110 are repeated for each input statement as it is read (unless an error is found
as indicated in steps 3 on page 110 and 4 on page 110).
IOCP:
1. Reads each statement from the input device.

2. Prints each statement on the output printer.
3. Checks the syntax of each statement for correct format. If IOCP finds a syntax error (such as a
missing parenthesis), IOCP prints an error message on the output printer after the statement
containing the error. IOCP stops checking the syntax of a statement when it encounters the first
syntax error in that statement. Therefore, if two or more syntax errors are in a single statement, the
second and following errors are found on subsequent runs of IOCP.
If IOCP finds an error in step 3, IOCP returns to step 1 on page 109 and performs only steps 1 on
page 109-4 for the remainder of the statements.
4. Checks the validity of each statement. For example, IOCP checks that a CNTLUNIT statement has
specified a channel path that has been previously defined in a CHPID statement. If an error is found,
IOCP prints an error message on the output printer after the statement containing the error.
If IOCP finds an error in step 4, IOCP returns to step 1 on page 109 and performs only steps 1 on
page 109-4 for the remainder of the statements.
5. Builds the appropriate identification, channel path, control unit, or I/O device record and enters the
record into the IOCDS in storage.
If more statements remain to be read, IOCP returns to step 1 on page 109.
After all the input statements have been processed and all the records have been entered into the
IOCDS in storage, IOCP performs step 6.
6. Generates the IOCDS and performs additional checking. If IOCP finds an error, IOCP sends an error
message and then stops.
After IOCP has successfully generated an IOCDS, IOCP performs Step 7.
7. Produces the IOCP configuration reports of the IOCDS.
For z/OS and z/VSE: If requested (by a WRTCDS option other than NO), IOCP performs step 8.
For VM: If requested (by a generation option of the IOCP command other than NOWRTCDS), IOCP
performs step 8.
8. Writes the generated IOCDS to the Support Element hard disk.
For z/OS: IOCP requires authorization before it writes an IOCDS. See “Authorizing an IOCDS write”
on page 112.
For z/VM: The control program checks your privilege class. You must have privilege class C to write
the IOCDS to the SE hard disk. You must have privilege class B to enable the IOCDS for dynamic I/O
configuration changes.
For z/VSE: This step requires permission from the system operator (through message ICP050D) before
IOCP writes to the SE hard disk.
Producing configuration reports

IOCP performs the following steps when you request configuration reports:
1. Reads one or more IOCDSs from the SE hard disk
2. Prints the IOCP configuration reports that you specify.
If IOCP encounters any errors during processing, it sends appropriate error messages to the output
printer. For examples of IOCP configuration reports, see Chapter 7, “IOCP configuration reports,” on page
151.

Chapter 3. Running the z/OS version of IOCP
This chapter describes the following topics:
v General information
v Authorization required to write an IOCDS
v How to build the IOCP module
v IOCP storage requirements
v JCL statements required to run IOCP functions (with examples)
v Return codes IOCP can set
v Methods you can follow to initially define the I/O configuration or to redefine an existing I/O
configuration.
General information
The z/OS version of the Input/Output Configuration Program (IOCP) runs as a job under control of
OS/390, z/OS, or z/OS.e. The IOCP program name is ICPIOCP. By running ICPIOCP, you can verify
IOCP input statements for CPCs that require the ICP version of IOCP. See Table 18 on page 287 to
determine which CPCs require this version of IOCP.
Hardware configuration definition (HCD) notes:

1. You do not need to run IOCP directly to define the I/O configuration to the I/O subsystem. However,
HCD does run IOCP when it needs to write an IOCDS.
2. If z/OS is a guest on VM, you cannot write an IOCDS to a CPC that is part of an S/390®
microprocessor cluster.
3. When HCD generates an IOCP input source file, it can include comments that HCD uses if the IOCP
source file is migrated into HCD. They have special meaning to HCD. These comments begin with
*$HCDC$ or *$HCD$. For more information, see OS/390 Hardware Configuration Definition User's Guide,
SC28-1848, or z/OS Hardware Configuration Definition User's Guide, SC33-7988.
4. When HCD generates an IOCP input source file without PCHIDs, the file must be migrated back into
HCD after the CHPID Mapping Tool adds PCHID numbers to the file. HCD can then write an IOCDS
or a new IOCP input file can then be created from HCD. Stand-alone IOCP can use this file to write
an IOCDS.
5. IOCP warning message ICP065I with return code 4 (RC=4) results when writing an IOCDS if your
I/O configuration contains an unsupported function or channel path type for the current EC level of
your CPC. This situation in turn results in the following HCD message.
CBDA676I IOCP completed with warning messages for <proc-id>.
See the IOCP output listing to view the IOCP warning messages.
Use JCL statements to run IOCP. In the EXEC statement, code options in the PARM keyword to direct
IOCP to do the following tasks:
v Generate an IOCDS:
– Read, validate, and check the syntax of input statements
– Generate an IOCDS based on the configuration defined in the statements
– Produce reports of the IOCDS
– When requested, write the generated IOCDS to the Support Element hard disk.
v Produce configuration reports:

– Read one or more IOCDSs from the Support Element hard disk
– Print IOCP configuration reports (see Chapter 7, “IOCP configuration reports,” on page 151).
Authorizing an IOCDS write

Writing an IOCDS requires authorization in either of the following ways:
v The z/OS system operator allows you by responding to message ICP050D
v A security product (for example, RACF) defines you as someone authorized to write an IOCDS.
The following conditions apply when you run IOCP directly to write an IOCDS or when you use the
Hardware Configuration Definition. (The HCD runs IOCP to build an IOCDS.)
v If the CBD.CPC.IOCDS resource of the FACILITY class exists in an installed security product (for
example, RACF), IOCP does not send message ICP050D to the z/OS operator. IOCP determines your
authorization to write an IOCDS from your authorization to update the CBD.CPC.IOCDS resource. If
the security administrator has authorized you to update this resource, IOCP writes the IOCDS. If you
are not authorized to update the resource, IOCP does not write the IOCDS.
v If the CBD.CPC.IOCDS resource of the FACILITY class does not exist in an installed security product,
IOCP sends message ICP050D. This message prompts the z/OS system operator to allow you to write
the IOCDS.
v The logical partition running IOCP must have I/O Configuration Control Authority enabled on the
Support Element.
Building the IOCP module

The IOCP program is on the SYS1.AOSC5 data set of the distribution library (DLIB). It is built when the
distribution libraries are loaded during the SMP product install procedure. During the SMP product
install procedure, IOCP is link-edited from the DLIB into SYS1.LINKLIB as an APF-authorized program,
authorization code AC=1, with the load module name ICPIOCP.
IOCP storage requirements

IOCP uses a variable amount of storage based on the type of IOCDS it is processing. To determine the
specific IOCP storage requirements for your CPC, see Table 18 on page 287. Specify a region size in the
REGION parameter of the JOB statement that is large enough to ensure sufficient space for the job or
specify 0 MB.
Coding the JCL EXEC statement for generating an IOCDS

In the EXEC statement, code the name of the IOCP program in the PGM parameter and the following
PARM options to generate an IOCDS. (The examples later in this chapter describe the JOB and DD
statements for an IOCP job.)
Note: PARM parameter defaults allow you to generate an IOCDS and to produce reports. However, the
defaults do not allow you to write an IOCDS to the Support Element hard disk.
IOCP processes all input statements and parameters. IOCP flags as errors any non-IOCP input and also
indicates errors found in IOCP statements and parameters. IOCP prints all input, including all comment
cards (all cards with an asterisk (*) in column 1 or .* in columns 1 and 2).
PARM=’WRTCDS=Ax|NO,LINECOUNT=n|55,CHECKCPC=YES|NO’
WRTCDS=
specifies whether or not IOCP writes an IOCDS to the support Element hard disk. All of the
WRTCDS options except NO cause IOCP to write an IOCDS. You can write an IOCDS only from an
authorized logical partition (that is, I/O Configuration Control Authority is enabled).
IOCP writes the IOCDS only if:
v No errors are encountered during IOCP processing
v The z/OS system operator grants permission in response to message ICP050D or the user
attempting to write the IOCDS is authorized by an installed security product (for example, RACF)
v The IOCDS is not write protected.
Request IOCP to write the IOCDS only after you have done the following:
v Checked the syntax of the input statements
v Generated an IOCDS based on the configuration defined in the statements
v Reviewed reports on the generated IOCDS
v Checked to see whether or not the generated IOCDS is correct.
To replace a write protected IOCDS, you must first remove write protection. Click Disable Write
Protection n under Options on the Input/Output Configuration window of the Support Element to
remove write protection.
HCD and write protection: You can use HCD to set and remove IOCDS write protection. For more
information, see z/OS Hardware Configuration Definition User's Guide, GC33-7988.
When IOCP writes an IOCDS to the Support Element hard disk, it overwrites the previous
configuration data in that IOCDS, even if that IOCDS is the active IOCDS. (The active IOCDS is the
one used for the last power-on reset or the last dynamic activation from the system control program
(SCP).)
The valid options for WRTCDS are Ax and NO.
Ax specifies that IOCP should write the Ax IOCDS (where x is 0, 1, 2, or 3) to the Support Element
hard disk.
NO specifies that IOCP should not write the IOCDS to the Support Element hard disk.
NO is the default if you do not code WRTCDS.
Code or use the default WRTCDS=NO when you test the input file and do not want to write to
the Support Element hard disk until you have successfully tested and checked the input file.
Also, you must code or use the default WRTCDS=NO when you run IOCP on a CPC that does
not require ICPIOCP. However, certain CPCs allow an IOCDS to be written in anticipation of
supporting ICPIOCP.
LINECOUNT=
n specifies the number of lines that IOCP is to print on a page of the IOCP output listing and
configuration reports, including the heading of the report. Code a decimal value in the range of
20-99.
55 is the default number of lines (including the heading) that IOCP prints on a page of the IOCP
output listing and configuration reports.
CHECKCPC=
YES
specifies that IOCP generates an IOCDS as follows:
v Based on the CPC on which IOCP is running if writing an IOCDS. IOCP ignores the SYSTEM=
keyword on the ID statement if specified.
v Based on the CPC on which IOCP is running if not writing an IOCDS and the SYSTEM=
keyword was not specified.
v Based on the SYSTEM= parameter if specified and IOCP is not writing an IOCDS.
Chapter 3. Running the z/OS version of IOCP 113

YES is the default.
NO specifies that IOCP writes the IOCDS in preparation of a CPC upgrade. IOCP generates an
IOCDS based on the SYSTEM= parameter on the ID statement and writes the IOCDS to the
current CPC. See “Machine rules” on page 285 to determine which SYSTEM= parameters are
valid with CHECKCPC=NO and which CPCs can have ICPIOCP write an IOCDS to them.
The IOCDS cannot be used to power-on reset the current CPC until it is upgraded to the type of
CPC that the IOCDS supports. For example, if IOCP generates an IOCDS for a 2084 CPC using
SYSTEM=(2084,1) and writes the IOCDS to a general purpose 2064 model 1xx, the IOCDS cannot
be used for power-on reset until the 2064 is upgraded to a 2084 model CPC.
A WRTCDS option other than NO is required. Also, the SYSTEM keyword on the ID statement
must be specified.
Attention: Use CHECKCPC=NO only in preparation of a CPC upgrade. After the upgrade, IBM
recommends that you write an IOCDS using CHECKCPC=YES at your earliest convenience. This
prevents the possibility of using the initial IOCDS for a second upgrade to a CPC, which will not
support the IOCDS and will fail power-on reset.
JCL examples of generating an IOCDS

Example 1
The following example shows the JCL statements used to check an IOCP input file. IOCP requires SYSIN
as the ddname for the input data set and SYSPRINT as the ddname for the output data set. Reports are
sent to SYSOUT.
//IOCP1 JOB REGION=0M,...
//TEST1 EXEC PGM=ICPIOCP,PARM=(’WRTCDS=NO’,
// ’LINECOUNT=60’)
//SYSIN DD DSNAME=USER.IOCD1.CARDS,DISP=SHR
//SYSPRINT DD SYSOUT=A,DCB=(RECFM=FA)
In the example:
IOCP1
specifies the JOB statement. Code parameters on the JOB statement that your installation requires.
To determine the region size to use, see “IOCP storage requirements” on page 112.
TEST1
specifies the EXEC statement and the program (ICPIOCP) to run. The PARM options are:
WRTCDS=NO
specifies not writing the IOCDS to the Support Element hard disk
LINECOUNT=60
specifies 60 lines of output on the reports IOCP produces.
SYSIN
specifies the DD statement that defines the input data set, which consists of statements with a
record length of 80 bytes.
SYSPRINT
specifies the DD statement that defines the output data set to receive messages and reports. (Note
that IOCP records contain ANSI device control characters and have a length of 133 bytes.)
Example 2
The following example shows the JCL statements to process an IOCP input file, to generate an IOCDS,
and to write the generated IOCDS to the A0 IOCDS on the Support Element hard disk. Reports of the
generated IOCDS are sent to SYSOUT.
IOCP requires the ddnames SYSIN for the input data set and SYSPRINT for the output data set.

//WRITE1 EXEC PGM=ICPIOCP,PARM=’WRTCDS=A0’
//SYSIN DD DSNAME=USER.IOCD2.CARDS,DISP=SHR
In the example:
IOCP2
specifies the JOB statement. Code parameters on the JOB statement required by your installation.
To determine the region size to use, see “IOCP storage requirements” on page 112.
WRITE1
specifies the EXEC statement and the program (ICPIOCP) to run. The PARM options are:
WRTCDS=A0
specifies writing the A0 IOCDS to the Support Element hard disk.
SYSIN
specifies the DD statement that defines the input data set, which consists of statements with a
record length of 80 bytes.
SYSPRINT
Coding the JCL EXEC statement for producing configuration reports

PARM options to produce configuration reports. (The example later in this chapter describes the JOB and
DD statements for an IOCP job.)
You must code the REPORT option to receive reports of an IOCDS.

PARM=’REPORT=(A|Ax[,Ax...]),LINECOUNT=n|55’
REPORT=
specifies that IOCP read one or more IOCDSs from the Support Element hard disk to produce
configuration reports. You can code more than one REPORT option. If you code only one option, you
can omit the parentheses.
Notes:
1. You cannot use ICPIOCP to produce reports for IOCDSs created by IYPIOCP, IZPIOCP, or
IXPIOCP.
2. An IOCDS can be read only from an authorized logical partition (that is, I/O Configuration
Control Authority is enabled).
The valid options for REPORT= are:
Ax specifies the Ax IOCDS, where x is 0, 1, 2, or 3.
A specifies all the Ax IOCDSs. See Table 17 on page 286 to determine the Ax IOCDSs your CPC
supports.
LINECOUNT=
n specifies the number of lines that IOCP is to print on a page of the IOCP output listing and
configuration reports, including the heading of the report. Code a decimal value in the range
20-99.
55 Is the default number of lines (including the heading) that IOCP prints on a page of the IOCP
output listing and configuration reports.

Conditions for coding the REPORT options
v You can code REPORT only if the IOCDSs to report were written with ICPIOCP; otherwise, IOCP ends
with an error.
v You cannot code REPORT with the WRTCDS or CHECKCPC option. You cannot read an input file and
generate an IOCDS (which includes reports of the generated IOCDS), and produce reports by means of
the REPORT option in the same job.
v You can code several REPORT options. Redundant or duplicate values are allowed. For example,
REPORT=(A0,A1,A2,A3) is the same as REPORT=A. Regardless of the order in which you specify the
options, the reports are always produced in the following order: A0, A1, A2, A3.
JCL example of producing configuration reports

The following example shows the JCL statements to produce IOCP reports of all the Ax IOCDSs read
from Support Element hard disk. Reports are sent to SYSOUT. Note that you do not need a DD statement
to access an IOCDS from the Support Element hard disk.
IOCP requires SYSPRINT as the ddname for the output data set.
//REPORT2 EXEC PGM=ICPIOCP,PARM=’REPORT=A’
In the example:
IOCP5
specifies the JOB statement. Code the parameters on the JOB statement that your installation
requires. To determine the region size to use, see “IOCP storage requirements” on page 112.
REPORT2
specifies the EXEC statement and the program (ICPIOCP) to run. The PARM option of
REPORT=A specifies reading all the Ax IOCDSs from the Support Element hard disk to produce
reports.
SYSPRINT
Return codes
When you run the z/OS version of IOCP, it returns one of the following codes at the completion of the
job:
Table 11. Return codes for z/OS version of IOCP
Return code
Decimal Hex Description and action
0 0 IOCP has completed the requested function without error and might have sent
one or more informational messages. If so, the informational messages in the
output listings or the reports describe the specific conditions.
2 2 IOCP has completed the requested function without error but has issued one
or more caution messages. The caution messages in the output listings or the
reports describe the specific conditions.
or more warning messages. The warning messages in the output listings or the
8 8 IOCP has encountered an error and has ended processing before completing
the requested function. The error message in the output listing describes the
specific condition.

Table 11. Return codes for z/OS version of IOCP (continued)
Return code
12 C IOCP has ended because the output data set DCB failed to open. IOCP was
unable to print messages or reports. Ensure that the DD statement for the
output data set is correctly specified.
16 10 IOCP has ended because it could not obtain the needed storage by means of
the GETMAIN statement. To determine the region size to use, see “IOCP
storage requirements” on page 112.
Methods of running the z/OS version of IOCP

This topic suggests the steps that you can use to run the z/OS version of IOCP to do the following:
v Define and check the I/O configuration data to prepare for a new CPC
v Redefine and replace an I/O configuration on an installed CPC.
Note: In z/OS, you use HCD to define and change your I/O configuration. HCD runs IOCP to write an
IOCDS.
Initial definition of I/O configuration data

The following example describes the steps that an installation can use to initially define and check the
I/O configuration data to prepare for the installation of a new CPC.
1. Determine, from your physical plan, the I/O configuration requirements that must be met to install
the CPC.
2. Use HCD on a running z/OS system to define your I/O configuration for the new CPC. HCD
validates the I/O configuration rules. When the definition is complete, build a production IODF.
3. If you are installing a new CPC, do one of the following:
v If the new CPC will be connected using a LAN to the CPC where HCD is running and that CPC is
able to remotely write an IOCDS, you can use HCD to write the new IOCDS to the CPC after it is
installed. HCD uses the APF-authorized ICPIOCP program to write an IOCDS. Inform the service
representative which IOCDS to use.
v If you are upgrading an existing CPC, you might be able to use HCD to write an IOCDS to your
current CPC in preparation for the upgrade. See “Machine rules” on page 285 to determine which
CPCs can write an IOCDS in preparation for an upgrade and which CPCs can have ICPIOCP write
an IOCDS to them. If you can write an IOCDS to your current CPC in preparation for upgrading it
to a new CPC, do so and inform the service representative which IOCDS to use. This is the
preferred method for writing the initial IOCDS when upgrading a CPC.
v If the new CPC will not be connected using a LAN to the CPC where HCD is running or you are
not upgrading or cannot write an IOCDS in preparation for the upgrade, use HCD to produce an
IOCP input file. Download the input file to diskette (if it is available) or USB flash memory drive. If
the file is too big to fit, compress it using a zip compatible program. Source from a diskette must be
in DOS file format or have been compressed. Source from a USB flash memory drive must be in
VFAT or DOS file format or have been compressed. Give the diskette or USB flash memory drive to
the service representative, who uses this input file and stand-alone IOCP to write the IOCDS for the
new CPC. Instead of a diskette or USB flash memory drive, an uncompressed IOCP source file can
be imported to the SE from an FTP location by the service representative. System z9 and zSeries
servers do not support USB flash memory drives for IOCDS source files. See “Creating an IOCDS
in a new CPC” on page 145 for an example using the stand-alone version of IOCP.
Note: See “Authorizing an IOCDS write” on page 112 for a description of how to authorize the
write of an IOCDS.

4. The service representative performs a power-on reset of the new CPC with the appropriate IOCDS.
5. The service representative performs an initial program load (IPL) for the z/OS system with the IODF
built under Step 2 on page 117.
Redefinition and replacement of I/O configuration data

The following example describes the steps the installation can use to replace the existing I/O
configuration data on a CPC.
1. Use HCD to update your I/O configuration for the hardware and software in a work IODF based on
your currently active production IODF.
2. Build a new production IODF from the updated work IODF.
3. If the changes can be dynamically performed, use HCD or the z/OS Activate system command to
dynamically change the I/O configuration in the Hardware System Area (HSA) and in the z/OS
system. The updated IODF now becomes the new active IODF. The details of dynamic I/O
reconfiguration are found in z/OS Hardware Configuration Definition Planning, GA22-7525.
4. Write the new IOCDS to the support Element of the CPC using HCD. HCD uses IOCP to write an
IOCDS. If the IOCDS slot is write protected on the Support Element, click Disable Write Protection
under Options on the Input/Output Configuration window of the Support Element.
Note: See “Authorizing an IOCDS write” on page 112 for a description of how to authorize the write
of an IOCDS.
5. Use HCD to switch the IOCDS to become the active one for the next power-on reset.
6. If the I/O configuration changes cannot be dynamically performed (see Step 3), perform a power-on
reset of the CPC using the new IOCDS. Perform an IPL of the z/OS system with the new production
IODF.
7. Test the new configuration following the procedures established for your site.

Chapter 4. Running the VM version of IOCP
This chapter describes the following topics:
v CMS IOCP commands that run these functions (with examples)
configuration
General information
The VM version of IOCP runs under the control of z/VM.
Start the VM version of IOCP in the CMS environment. The CMS IOCP command runs IOCP and has
options you can use to request the following IOCP functions:
1. Generate an IOCDS.
IOCP reads, validates, and checks the syntax of input statements. It generates an IOCDS based on the
configuration defined in the statements. It produces reports of the IOCDS; and, when requested, IOCP
writes the generated IOCDS to the Support Element hard disk.
2. Produce configuration reports.
IOCP reads one or more IOCDSs from the Support Element hard disk and prints the IOCP
Hardware configuration definition (HCD) notes

1. When using HCD to define the I/O configuration for VM, you need not run IOCP directly to define
the I/O configuration to the I/O subsystem. However, HCD does run IOCP when it must write an
IOCDS. HCD users: Always use HCD (CBDSIOCP EXEC) to write an IOCDS from VM.
2. When HCD generates an IOCP input source file, it can include comments that HCD uses if the IOCP
source file is migrated into HCD. They have special meaning to HCD. These comments begin with
*$HCDC$ or *$HCD$. For more information, see z/VM Dynamic I/O Configuration Planning and
Administration, SC24-6044.
3. When HCD generates an IOCP input source file without PCHIDs, the file must be migrated back into
HCD after the CHPID Mapping Tool adds PCHID numbers to the file. HCD can then write an IOCDS
or a new IOCP input file can then be created from HCD, which a stand-alone IOCP can use to write
an IOCDS.
4. IOCP warning message ICP065I with return code 4 (RC=4) results when writing an IOCDS if your
I/O configuration contains an unsupported function or channel path type for the current EC level of
your CPC. This situation in turn results in the following HCD message.
CBDA676I IOCP completed with warning messages for <proc-id>.
See the IOCP output listing to view any IOCP warning messages.

IOCP uses a variable amount of storage based on the type of IOCDS it is processing. However, to
determine the specific IOCP storage requirements for your CPC, see Table 18 on page 287. The IOCP
command requires the storage size documented in the table, which is in addition to the storage CMS
requires.

The IOCP command
The format of the IOCP command is shown in the following diagram. The options are explained in the
text that follows the format.
IOCP [fn] [(options...[)]]
Program Options:
ICP
IOCDS Generation Options:

CHECKCPC NOWRTCDS
NOCHKCPC WRTAx
Configuration Report Options:

RPTAx
RPTA
Listing Control Options:

LINECOUN (55) DISK
LINECOUN (nn) PRINT
NOPRINT
System Terminal Options:

TERMINAL WARNING
NOTERM NOWARN
IOCP CMS Utility Options:

NODYN
DYN
fn Specifies the file name of the input file of IOCP statements or of the output listing file, depending on
the options you specify on the command line. A file name is required only when input is from a disk
or output is to a disk.
v When you specify IOCDS generation options, fn is the following:
– The file name of the input file that contains the IOCP statements.
– The file name of the output file.
If IOCP finds no errors, the output file contains the configuration reports of the IOCDS that
IOCP generated from your input and a list of your input. If IOCP finds errors, the output file
contains only a list of the input file and any errors that IOCP finds.
v When you specify configuration report options, fn is the following:
– The file name of the output file.
The output file contains the configuration reports of the IOCDSs you request from the Support
Element.
The IOCP CMS utility requires the IOCP input file to be on a read/write disk unless the NODYN option
is specified and the IOCP input file is not dynamic-capable.
IOCP issues, by default, the following FILEDEF commands for the SYSIN (input to IOCP) and SYSPRINT
(output from IOCP) ddnames:
FILEDEF SYSIN DISK fn IOCP * (RECFM FB LRECL 80 BLOCK 800
FILEDEF SYSPRINT DISK fn LISTING fm (RECFM FBA LRECL 133 BLOCK 1330
IOCP obtains fn from the file name entered in the IOCP command.
IOCP obtains fm by checking the disk that contains the input file. If the disk is read/write, IOCP writes
back to the same disk. However, if the disk is an extension to a disk, IOCP writes to the parent disk. If
neither of these cases is true, IOCP writes to the A-disk (provided it is read/write).

Because the default FILEDEF commands specify DISK, you must use a file name in the IOCP command
unless you override those default FILEDEF commands. If you want to override a default FILEDEF
command, issue your own FILEDEF command before running IOCP. The SYSPRINT FILEDEF is
automatically overridden if you specify PRINT or NOPRINT in the IOCP command. Note also that if you
specify a SYSPRINT FILEDEF, the IOCP command uses a system terminal option of NOTERM, which
cannot be overridden.
The input file type is, by default, IOCP. If the input file has a file type other than IOCP, issue a FILEDEF
command to override the default file type.
Program options
A program option designates the IOCP program for which the input IOCP statements were written.
The syntax of input IOCP statements can be checked on any CPC in which the CMS IOCP command is
valid. Syntax checking can be performed when the CPC for which the input IOCP statements were
written is not available.
The valid program options are as follows:

ICP
Specifies that the ICP version of IOCP is to process the command options and input file to generate
an IOCDS. ICP indicates the first three characters of the ICPPCNTL IOCP module and to the z/OS
program name ICPIOCP. See Table 18 on page 287 to determine which CPCs require this version of
IOCP.
| The ICP program option is the default for the IOCP command on 3907, 3906, 2965, 2964, 2828, 2827,
| 2818, 2817, 2098, 2097, 2096, 2094, 2086, and 2084 CPCs.
xxx
Three-character program option other than ICP. It is associated with CPCs that do not require the ICP
version of IOCP.
For example, the IYP program option runs the version of IOCP that is required on 2064 and 2066
CPCs. The IYP program option is the default for the IOCP command on 2064 and 2066 CPCs.
The IZP program option runs the version of IOCP that is required on 9021, 9121, 9672, 2003, 3000,
and 7060 CPCs that support the ESCON architecture and multiple image facility (MIF). The IZP
program option is the default for the IOCP command on 9672, 2003, 3000, and 7060 CPCs.
The IXP program option runs the version of IOCP that is required on ES/3090, 9021, 9121, and 9221
CPCs that support the ESCON Architecture, but not the multiple image facility (MIF). The IXP
program option is the default for the IOCP command on 9021, 9121, and 9221 CPCs.
IOCP programs are designed to generate IOCDSs for specific CPCs. However, some IOCP programs can
generate an IOCDS for a CPC it was not designed to support. For example, if IOCP writes an IOCDS
with the ICP program option on a CPC that does not support ICP IOCP, a power-on reset failure occurs if
the IOCDS is used before the CPC is upgraded to a CPC that supports ICP IOCP.
IOCP programs are also designed to report on IOCDSs for specific CPCs. An IOCP program can report
only on IOCDSs that were written by the same IOCP program. For example, if IOCP writes an IOCDS
with the ICP program option, only the ICP IOCP program can print a report. Errors occur when an IOCP
program other than the one that generated the IOCDS tries to report on an IOCDS.
IOCDS generation options

IOCP writes an IOCDS only if:
v No errors are encountered during IOCP processing.
Chapter 4. Running the VM version of IOCP 121

v The user issuing the IOCP command has the correct privilege class. See Note 1 on page 125 in “IOCP
command – usage notes for VM” on page 125.
v The IOCDS is not write-protected.
v You specify a write IOCDS option WRTAx.
Use a write option WRTAx after you have:

v Checked to see whether the I/O configuration generated by IOCP corresponds to the I/O configuration
in the system configuration file (SYSTEM CONFIG) on the parm disk or the obsolete real I/O
configuration file (HCPRIO) in the CP nucleus. While IOCP must define the entire I/O configuration,
only the devices that VM cannot sense during IPL must be defined in the SYSTEM CONFIG or
HCPRIO files.
If you want to replace an IOCDS that is write-protected, you must also remove the write protection. Use
Disable Write Protection under Options on the Input/Output Configuration window of the Support
Element to remove write protection.
When IOCP writes the generated IOCDS to the Support Element hard disk, the previous configuration
data in that IOCDS is overlaid and lost, even if that IOCDS is the active IOCDS. (The active IOCDS is the
one used for the last power-on reset or the last dynamic activation from the system control program
(SCP).)
NOWRTCDS
NOWRT
Specifies that IOCP is not to write an IOCDS to the Support Element hard disk. This option is a
default.
Use the NOWRTCDS option in either of the following situations:
v You are checking the syntax of the input file and do not want to write to the Support Element hard
disk until you have successfully checked the input file.
v You are using a CPC that does not require the IOCP that this publication describes.
However, certain CPCs permit an IOCDS to be written in anticipation of supporting ICP IOCP.
When you specify the NOWRTCDS option and IOCP finds no errors, the output file contains the
configuration reports of the IOCDS that IOCP generates from your input, along with a list of your
input file. If IOCP finds errors, your output file contains only a list of your input file and any errors
that IOCP finds.
WRTAx
WAx
Specifies that IOCP is to write the Ax (where x is 0, 1, 2, or 3) IOCDS to the Support Element hard
disk. Privilege class C is required to write an IOCDS.
Note: An IOCDS can be written only from an authorized logical partition (that is, I/O Configuration
CHECKCPC
Specifies that IOCP generates an IOCDS as follows:
v Based on the CPC on which IOCP is running if not writing an IOCDS and the SYSTEM= keyword
was not specified.

This option is the default.
NOCHKCPC
Specifies that the IOCDS is being written in preparation of a CPC upgrade. IOCP generates an IOCDS
based on the SYSTEM= parameter on the ID statement and writes the IOCDS to the current CPC. See
“Machine rules” on page 285 to determine which CPCs can write an IOCDS with NOCHKCPC and
which CPCs can have an IOCDS written to them by ICP IOCP.
The IOCDS cannot be used to power-on reset the current CPC until it is upgraded to the type of CPC
supported by the IOCDS. For example, if IOCP generates an IOCDS for a 2084 CPC using
SYSTEM=(2084,1) and writes the IOCDS to a general purpose 2064 model 1xx, the IOCDS cannot be
used for power-on reset until the 2064 is upgraded to a 2084 model CPC.
A WRTAx or WAx option is required. Also, the SYSTEM keyword on the ID statement must be
specified.
Attention:
Use NOCHKCPC only in preparation of a CPC upgrade. After the upgrade, write an IOCDS using
CHECKCPC at your earliest convenience. Doing so precludes the possibility of using the initial
IOCDS for a second upgrade to a CPC, which does not support the IOCDS and fails power-on reset.
Configuration report options

The valid options for reading an IOCDS and producing configuration reports are RPTAx and RPTA.
The user issuing the IOCP command to produce configuration reports must have the correct privilege
class. See Note 1 on page 125 in “IOCP command – usage notes for VM” on page 125.
Note: You cannot use the ICP version of IOCP to produce reports for IOCDSs created by the IYP, IZP, or
IXP versions of IOCP.
RPTAx
RAx
Specifies that IOCP reads the Ax (where x is 0, 1, 2, or 3) IOCDS from the Support Element hard disk
and generates a configuration report of that data set. You can repeat this option to generate reports
on multiple IOCDS files. Privilege class C is required to read an IOCDS.
Note: An IOCDS can be read only from an authorized logical partition (that is, I/O Configuration
RPTA
RA Specifies that IOCP is to read all the Ax IOCDSs from the Support Element hard disk and is to
generate configuration reports of those data sets. See Appendix F, “Machine limits and rules,” on
page 285 to determine the Ax IOCDSs your CPC supports. Privilege class C is required to read an
IOCDS.
Note: An IOCDS can be read only from an authorized logical partition (that is, I/O Configuration
Listing control options

LINECOUN (55)
LINECOUN (nn)
Specifies the number of lines (from 20 to 99) to print on each page including the report heading. The
default is LINECOUN (55). You must specify the number of lines if you use this option.
DISK
DI Specifies that IOCP is to place the output listing on a minidisk. DISK is a default option.
PRINT

PR Specifies that IOCP is to send the output listing to the printer.
NOPRINT
NOPR
Specifies that IOCP is not to generate any output listing.
System terminal options

TERMINAL
TERM
Specifies that IOCP is to write diagnostic information from the IOCP output to your terminal.
Diagnostic information consists of either of the following:
v An IOCP message (see Chapter 8, “IOCP messages,” on page 171) and, preceding this message, the
nonblank line that caused the diagnostic message
v A display of the total message count for the IOCP output.
TERMINAL is a default option.
NOTERM
Specifies that IOCP is to suppress the display of diagnostic messages at your terminal.
WARNING
WARN
Specifies that IOCP is to write IOCP warning and caution messages to the terminal. This option is
valid only if you specify the TERMINAL option. WARNING is a default option.
NOWARN
Specifies that IOCP is to suppress the writing of warning and caution messages at your terminal.
IOCP CMS utility options

These options are valid only for the IOCP CMS utility. If an IOCP CMS utility option is specified, the
utility processes the option and then calls the IOCP command with all specified options except the utility
options. The IOCP command does not accept the IOCP utility options.
NODYN
Specifies that the IOCP CMS utility is not to enable the dynamic I/O configuration capability of any
resulting IOCDS. This option is the default for CPCs that do not have a fixed storage size for the
Hardware System Area (HSA). See Table 16 on page 285 to determine if your CPC has fixed HSA.
NODYN disables any dynamic-capability of an IOCP input file. NODYN requires the IOCP input file
to be on a read/write disk unless the file is not dynamic-capable. Specify NODYN when you are not
writing an IOCDS and do not have privilege class B.
Note: You cannot specify utility option NODYN along with the MAXDEV keyword or reserved
logical partitions on the RESOURCE statement.
DYN
Specifies that the IOCP CMS utility is to enable the dynamic I/O configuration capability of any
resulting IOCDS. This option is the default for CPCs that have a fixed storage size for the Hardware
System Area (HSA). See Table 16 on page 285 to determine if your CPC has fixed HSA. The DYN
option requires the following:
v Privilege class B
v An IOCP input file on a read/write disk.
Restrictions
CMS/DOS does not recognize the IOCP command.

IOCP command – general usage notes
This topic lists some general notes to keep in mind when you use the IOCP command. Some aspects of
using the command, though, vary with the operating system on which you run IOCP. For usage notes
specific to your operating system, see the topics that immediately follow this topic.
Notes:
1. Some IOCP command options cancel other IOCP command options. For this reason, when you code
the IOCP command, do not combine IOCDS generation options with configuration report options.
(However, IOCP allows multiple configuration-report options.)
If you inadvertently code conflicting options in the same IOCP command, IOCP uses the last entered
of those conflicting options; no warning message is sent.
2. The write option WRTAx, is incorrect if the data set that you want to replace is write-protected. (To
remove write protection, click Disable Write Protection under Options on the Input/Output
Configuration window of the Support Element.)
3. The TERMINAL and WARNING options are not active if you provide your own SYSPRINT FILEDEF
command.
IOCP command – usage notes for VM

This topic lists some notes to keep in mind when you use the IOCP command.
Notes:
1. You must have the correct privilege class. The control program checks your privilege class when IOCP
attempts to read or write an IOCDS. You must have privilege class C to read or write an IOCDS; you
must also have privilege class C to avoid IOCP warning message ICP068I when performing a deck
verification (option NOWRTCDS). You must have privilege class B if you specify IOCP utility option
DYN.
If you try to read from an IOCDS or write an IOCDS and you do not have the required privilege
classes, you receive the following messages:
ICP402I IOCP RECEIVED AN UNEXPECTED RESPONSE CODE OF 01F0 HEX
ATTEMPTING READ OR WRITE OF AN IOCDS
ICP051I IOCP VERSION nn.nn.nnn TERMINATED. CODE=xx
Note: You can also receive the previous messages if you try to read or write an IOCDS when another
user is reading or writing an IOCDS. The messages are printed on the SYSPRINT device. IOCP then
stops processing.
If you try to read or write an IOCDS or perform a deck verification (option NOWRTCDS) and you do
not have the required privilege classes, you receive the following message:
ICP068I **WARNING** IOCP WAS UNABLE TO DETERMINE IF THE CONFIGURATION
IS USABLE ON THE EC LEVEL OF THIS PROCESSOR (CODE 01F00000)
If you specify the DYN IOCP utility option and do not have the required privilege class, you receive
the following message from the IOCP utility:
Error occurred generating hardware token.
CP QUERY TOKEN command was not recognized.
IOCP module has not been invoked.
2. For z/VM 4.4.0 and prior releases, if you do not have CP APAR VM63426 installed, you receive the
following message:
ICP068I **WARNING** IOCP WAS UNABLE TO DETERMINE IF THE CONFIGURATION
IS USABLE ON THE EC LEVEL OF THIS PROCESSOR (CODE 40F00000)
3. You receive the following error messages and IOCP processing stops if you try to update an IOCDS or
request a configuration report for an IOCDS while IOCP is running on a CPC that does not require
the version of IOCP described in this publication:

ICP400I READ/WRITE OF IOCDS IS INVALID ON THIS PROCESSOR (nnnn)
You receive similar error messages if you try to update an IOCDS or request a configuration report
for an IOCDS on a CPC that supports ICP IOCP and a program option other than ICP was specified.
4. If you and another user attempt to read the same IOCDS at the same time, or if you and another user
attempt to write the same IOCDS at the same time, you or the other user receive the following
messages:
ICP405I IOCP RECEIVED AN UNEXPECTED RETURN CODE OF 09 OR 65
HEX ATTEMPTING READ OR WRITE OF AN IOCDS
Wait several minutes and again attempt to read or write that IOCDS by issuing another IOCP
command.
5. You must define your I/O configuration for the CPC using IOCP statements. When you perform an
initial program load (IPL) of VM, VM creates its software I/O configuration by sensing the devices in
the hardware I/O configuration. If all the devices return all of the information VM needs, VM
completes its software I/O configuration. If not, you must specify this information in the system
configuration file or the obsolete real I/O configuration file.
6. If you try to write an IOCDS and either the IOCP abnormally ends or you issue the CMS command
HALT EXECUTION (HX) while IOCP is running, you prevent another authorized user from using an
IOCDS. Under these circumstances, the system might have locked the IOCDS for you. If the other
user tries to write to any IOCDS, the following messages are sent:
ICP401I READ/WRITE IOCDS FACILITY TEMPORARILY BUSY (RC=4). RETRY LATER
The xx is a code that indicates IOCP encountered an error while trying to write the locked IOCDS.
If the other user attempts to read the IOCDS that has been locked, the following messages are
displayed to that user:
ICP403I LEVEL Ax IOCDS IS INVALID
The xx is a code that indicates IOCP encountered an error while trying to read the locked IOCDS.
These messages are sent each time the other user tries to run IOCP until you do one of the following:
v Rerun IOCP to completion, successfully writing the locked IOCDS to the Support Element.
v Send a SYSTEM RESET command to reset your virtual machine.
v Send a LOGOFF command to log off your virtual machine.
The other user can then access any of the IOCDSs.
Examples of generating an IOCDS

Example 1
This example shows the IOCP command options that check an IOCP input file. IOCP displays any
messages at your terminal. The output file MIFILE LISTING contains the configuration reports for the
configuration described by the input file MIFILE IOCP.
IOCP MIFILE (NOWRTCDS LINECOUN (60)
MIFILE
Is the name of the CMS file (file type IOCP) that contains IOCP statements.

NOWRTCDS
Specifies that IOCP does not write the IOCDS to the Support Element hard disk.
LINECOUN (60)
Specifies that IOCP prints 60 lines of output on each page of the configuration reports.
Example 2
This example shows the IOCP command options that perform the following tasks:
v Process an IOCP input file
v Generate an IOCDS
v Write the generated IOCDS to the A0 slot on the Support Element hard disk.
IOCP NEWIOCDS (WRTA0
NEWIOCDS
Is the name of the CMS file (file type IOCP) that contains IOCP statements.
WRTA0
Specifies that IOCP is to write the IOCDS generated to the A0 slot on the Support Element hard disk.
Examples of producing a configuration report

This example shows the IOCP command options used to produce a report of every IOCDS on the CPC.
Use these reports to analyze the present I/O configuration of your installation.
IOCP IOCDSRPT (RPTA LINECOUN (65) DISK NOTERM
IOCDSRPT
Is the file name of the output file that contains the configuration reports. The file type is LISTING.
RPTA
Specifies that IOCP creates a configuration report for every IOCDS (A0, A1, A2, and A3 IOCDSs).
LINECOUN (65)
Specifies that IOCP prints 65 lines of output on each page of a configuration report.
DISK
Specifies that IOCP sends its output to disk storage.
NOTERM
Specifies that IOCP does not display any diagnostic messages at the terminal.
Return codes
When you run the VM version of IOCP, IOCP returns one of the following codes at the completion of the
job:
Table 12. VM version of IOCP
Return code Description and action
(decimal)
0 IOCP has completed the requested function without error, and might have sent one or more
informational messages. If so, the informational messages in the output listings or the
2 IOCP has completed the requested functions without error but has sent one or more caution
messages. The caution messages in the output listings or the reports describe the specific
conditions.
4 IOCP has completed the requested functions without error but has sent one or more warning
messages. The warning messages in the output listings or the reports describe the specific
conditions.

Table 12. VM version of IOCP (continued)
Return code Description and action
(decimal)
8 IOCP has encountered an error and has ended processing before completing the requested
function. The error message in the output listing describes the specific condition.
12 IOCP has ended because the output data set DCB failed to open. IOCP was unable to print
messages or reports. Ensure that the FILEDEF command for the output data set is correct.
You can query the data definitions for all files by issuing the CMS FILEDEF command without
operands.
16 IOCP has ended because it was unable to obtain enough storage. To determine the necessary
amount of storage, see “IOCP storage requirements” on page 112. Define more virtual
storage, perform an IPL of CMS, and issue the IOCP command again.
In addition to the return codes listed, the CMS-IOCP interface module, DMSICP, sends its own error
messages with corresponding return codes for each message. For the return codes generated with error
messages, see “IOCP messages (DMSICP prefix)” on page 233.
Methods of running the VM version of IOCP

This topic suggests ways the installation can run the VM version of IOCP to:
1. Initially define and validate the I/O configuration data for a new CPC
2. Later redefine and replace the I/O configuration on an installed CPC.

Before installing the CPC, you can create an IOCP input file on an existing system. The service
representative uses this input file during the installation of the new CPC unless you are upgrading a CPC
and were able to write an IOCDS in preparation of the upgrade. To prepare an input file, follow these
steps:
1. Determine, from your physical plan, the I/O configuration requirements that must be met to install
the CPC.
2. Prepare an input file of IOCP statements that define your I/O configuration and system requirements.
The service representative uses this file when installing the CPC. For a sample input file containing
IOCP statements, see Figure 31 on page 241.
3. Check the IOCP statement input by running the IOCP command under CMS. Code NOWRTCDS in
the IOCP command. Running this command checks the syntax of the IOCP statements and checks the
configuration reports for the wanted configuration data.
Note: When you perform an initial program load (IPL) of VM, VM creates its software I/O
configuration by sensing the devices in the hardware I/O configuration. If all the devices return all of
the information VM needs, VM completes its software I/O configuration. If not, you must specify this
information in the system configuration file or the obsolete real I/O configuration file. Ensure that
this information corresponds to the information you code in the IOCP statements. Figure 32 on page
241 contains sample SYSTEM CONFIG statements that correspond to some of the IOCP statements in
Figure 31 on page 241. Figure 33 on page 242 contains a sample HCPRIO file that corresponds to the
sample IOCP input file shown in Figure 31 on page 241.
4. If there are any errors in the IOCP statements, correct the errors and rerun IOCP (Step 3).
5. Analyze the IOCP configuration reports (produced in Step 3) to ensure that the I/O configuration you
want is defined. If necessary, revise the IOCP statements and rerun IOCP.
6. If you are upgrading to a CPC whose SYSTEM= value supports IOCDS generation option
NOCHKCPC, write the IOCDS for the new CPC on your current CPC using NOCHKCPC. See
“Machine rules” on page 285 to determine which CPCs can write an IOCDS in preparation for an
upgrade and which CPCs can have an IOCDS written to them by ICP IOCP. This method is the

preferred method for writing the initial IOCDS when upgrading a CPC. Inform the service
representative which IOCDS to use after the upgrade. Skip Step 7.
If you are upgrading to a CPC whose SYSTEM= value does not support parameter NOCHKCPC or if
this installation is a new installation, continue with Step 7.
7. Download the input file to diskette (if it is available) or USB flash memory drive. If the file is too
large to fit, compress it using a zip-compatible program. Source from a diskette must be in DOS file
format or have been compressed. Source from a USB flash memory drive must be in VFAT or DOS
file format or have been compressed. Instead of a diskette or USB flash memory drive, an
uncompressed IOCP source file can be imported to the SE from an FTP location by the service
representative. System z9 and zSeries servers do not support USB flash memory drives for IOCDS
source files.
Note: During the physical installation of the CPC, the service representative uses your validated input
file (with the stand-alone version of IOCP) to configure and test the I/O configuration for the new
CPC. See “Creating an IOCDS in a new CPC” on page 145 for an example using the stand-alone
version of IOCP.

The following steps describe how your installation can replace the existing I/O configuration data on a
CPC:
1. Update the input file that contains the IOCP statements to reflect your new I/O configuration.
Figure 31 on page 241 contains a listing of a typical IOCP input file.
2. Run the VM version of IOCP to check the input file you updated in Step 1. Use the NOWRTCDS
options in the CMS IOCP command.
3. Correct any errors in the IOCP statements and rerun IOCP (Step 2).
4. Analyze the IOCP configuration reports (produced in Step 2) to ensure that the I/O configuration
that you want is defined. If necessary, revise the IOCP statements and rerun IOCP (Step 2).
5. Run the IOCP program by issuing the IOCP command with a write option WRTAx. This option
writes the I/O configuration to the specified IOCDS on the Support Element. All users must have
privilege class C to write. You must have privilege class B if you specify IOCP utility option DYN.
Note: Continue with Steps 6-11 when you want to use the new IOCDS created in Step 5. Perform
these steps at a convenient time, because they disrupt installation operations.
6. If necessary, revise the system configuration file or the real I/O configuration file to reflect the
changes made by IOCP. The real I/O configuration file is HCPRIO. The system configuration file is
SYSTEM CONFIG.
Figure 32 on page 241 contains sample SYSTEM CONFIG statements that correspond to some of the
IOCP statements in Figure 31 on page 241. Figure 33 on page 242 contains a sample HCPRIO list that
corresponds to the IOCP list in Figure 31 on page 241.
7. Generate the new VM system.
8. Shut down VM.
9. Perform a power-on reset of the CPC with the new IOCDS specified as the source of your
configuration data. Click Disable Write Protection under Options on the Input/Output
Configuration window of the Support Element.
10. Load (IPL) VM.
11. Test the new configuration. Follow the procedures established for your site.
12. Save the file you used in Step 2 to provide a backup.

Chapter 5. Running the z/VSE version of IOCP
This chapter describes the following:
v JCL statements required to run IOCP functions (with examples)
configuration.
General information
The z/VSE version of the input/output configuration program (IOCP) runs as a job under control of
z/VSE. The IOCP program name is ICPIOCP. By running ICPIOCP, you can verify IOCP input statements
for CPCs that require the ICP version of IOCP. See Table 18 on page 287 to determine the CPCs that
require ICPIOCP.
In z/VSE, use JCL statements to run IOCP. In the EXEC statement, code options in the PARM keyword to
override IOCP defaults. Use z/VSE to direct IOCP to do the following tasks:
v Generate an IOCDS:
– Read, validate, and check the syntax of input statements
– Generate an IOCDS based on the configuration defined in the statements
– Produce reports of the IOCDS
– When requested, write the generated IOCDS to the Support Element hard disk.
v Produce configuration reports:
– Read one or more IOCDSs from the Support Element hard disk
– Print IOCP configuration reports (see Chapter 7, “IOCP configuration reports,” on page 151).

IOCP uses a variable amount of storage based on the type of IOCDS it is processing. Use a partition size
large enough (at least 100 MB) to ensure sufficient space for the job. However, to determine the specific
IOCP storage requirements for your CPC, see Table 18 on page 287.
Coding the JCL EXEC statement for generating an IOCDS

In the EXEC statement, code the name of the IOCP program in the PGM parameter and code the
following PARM options to generate an IOCDS. (The examples later in this chapter describe the JOB and
ASSGN statements for an IOCP job.)
Note: You can use PARM parameter defaults to generate an IOCDS and to produce reports. However, the
defaults do not allow you to write to the Support Element hard disk.
PARM=’WRTCDS=Ax|NO,LINECOUNT=n,CHECKCPC=YES|NO’
WRTCDS=
Specifies whether IOCP writes an IOCDS to the Support Element hard disk. All WRTCDS options
except NO cause IOCP to write an IOCDS. You can write an IOCDS only from an authorized logical
partition (that is, I/O Configuration Control Authority is enabled).
Note: If you run z/VSE as a guest under z/VM, writing an IOCDS is not supported.

IOCP writes the IOCDS only if:
v No errors are encountered during IOCP processing
v The system operator grants permission in response to message ICP050D
v The IOCDS is not write protected.
Request IOCP to write the IOCDS only after you have done the following tasks:
v Checked to see whether the generated IOCDS is correct.
To replace a write-protected IOCDS, click Disable Write Protection under Options on the
Input/Output Configuration window of the Support Element.
When IOCP writes an IOCDS to the Support Element hard disk, it overwrites the previous
configuration data in that IOCDS, even if that IOCDS is the active IOCDS. (The active IOCDS is the
one used for the last power-on reset.)
The valid options for WRTCDS are Ax and NO.
Ax Specifies that IOCP writes the Ax IOCDS (where x is 0, 1, 2, or 3) to the Support Element hard
disk.
NO Specifies that IOCP does not write the IOCDS to the Support Element hard disk.
NO is the default if you do not code WRTCDS.
Code or use the default WRTCDS=NO when you test the input file and do not want to write to
the Support Element hard disk until you have successfully tested and checked the input file.
Also, you must code or use the default WRTCDS=NO when you run IOCP on a CPC that does
not require the IOCP this publication describes. However, certain CPCs permit you to write an
IOCDS in anticipation of supporting ICP IOCP.
LINECOUNT=n
Specifies the number of lines that IOCP prints on a page of the IOCP output listing and configuration
reports, including the heading of the report. Code a decimal value in the range of 20-99.
If you do not specify LINECOUNT, use the equivalent z/VSE command SET LINECT=n.
CHECKCPC=
YES
Specifies that IOCP generate an IOCDS as follows:
v Based on the CPC on which IOCP is running if not writing an IOCDS and the SYSTEM=
keyword was not specified.
YES is the default.
NO Specifies writing the IOCDS in preparation of a CPC upgrade. IOCP generates an IOCDS based
on the SYSTEM= parameter on the ID statement and writes the IOCDS to the current CPC. See
“Machine rules” on page 285 to determine which SYSTEM= parameters are valid with
CHECKCPC=NO and which CPCs can have ICPIOCP write an IOCDS to them.
The IOCDS cannot be used to power-on reset the current CPC until it is upgraded to the type of
CPC the IOCDS supports. For example, if IOCP generates an IOCDS for a 2084 CPC using
SYSTEM=(2084,1) and writes the IOCDS to a general purpose 2064 model 1xx, the IOCDS cannot
be used for power-on reset until the 2064 is upgraded to a 2084 model CPC.

A WRTCDS option other than NO is required. Also, the SYSTEM keyword on the ID statement
must be specified.
Attention: Use CHECKCPC=NO only in preparation of a CPC upgrade. After the upgrade,
write an IOCDS using CHECKCPC=YES at your earliest convenience. Doing so prevents the
possibility of using the initial IOCDS for a second upgrade to a CPC that does not support the
IOCDS and fails power-on reset.
JCL examples of generating an IOCDS

Example 1
The following example shows the JCL statements to check an IOCP input file embedded in the job. IOCP
requires SYSLST for the output data set and sends reports there.
// JOB IOCP1
// ASSGN SYSLST,cuu
// ON $RC GE 8 GOTO $EOJ
// EXEC PGM=ICPIOCP,SIZE=AUTO
(Imbed your IOCP input statements here)
/*
/&
In the example:
JOB
Specifies the JOB statement.
SYSLST
Specifies the printer to receive messages and reports. You must assign SYSLST to a physical printer or
to a VSE/POWER spooled printer. (IOCP records contain ANSI device control characters and have a
length of 133 bytes.)
ON $RC GE 8 GOTO $EOJ
Specifies that if IOCP ends with an error (return code of 8 or greater), job processing ends.
EXEC
Specifies the EXEC statement with the following options coded:
PGM=ICPIOCP
Specifies that you want to run the ICPIOCP program.
SIZE=AUTO
Specifies that the IOCP program uses the remaining partition size for GETVIS.
Example 2
This example shows the JCL statements used to do the following tasks:
v Check an input file
v Generate an IOCDS
v Write the generated IOCDS to the A0 IOCDS slot on the Support Element.
IOCP requires SYSLST for the output data set and sends reports there.
// JOB IOCP2
// ASSGN SYSLST,cuu
// ON $RC GE 8 GOTO label
// EXEC PGM=ICPIOCP,SIZE=AUTO,PARM=’WRTCDS=A0’
* $$ SLI ICCF=(membername),LIB=(lib)
/*
/. label
/&
Chapter 5. Running the z/VSE version of IOCP 133

In the example:
JOB
SYSLST
ON $RC GE 8 GOTO label
Specifies that if IOCP ends with an error (return code of 8 or greater), job processing continues at
label.
EXEC
PGM=ICPIOCP
SIZE=AUTO
PARM='WRTCDS=A0'
Specifies that IOCP write the A0 IOCDS to the hard disk of the Support Element.
ICCF=(membername)
Specifies the name of the IOCP source member of a VSE/ICCF library.
LIB=(lib)
Specifies the name of the VSE/ICCF-controlled library to search for the named SLI member.
Example 3
The following example shows the JCL statements to check an IOCP input file. IOCP requires SYSLST for
the output data set and sends reports there.
// JOB IOCP1
// ASSGN SYSLST,cuu
// ASSGN SYSIPT,cuu
// ON $RC GE 8 GOTO $EOJ
// EXEC PGM=ICPIOCP,SIZE=AUTO
/*
/&
In the example:
JOB
SYSLST
SYSIPT
Specifies the device that contains the input data set, which consists of statements with a record length
of 80 bytes. You can assign SYSIPT to a disk, tape, or card reader.
ON $RC GE 8 GOTO $EOJ
Specifies that if IOCP ends with an error (return code of 8 or greater), job processing ends.
EXEC

PGM=ICPIOCP
SIZE=AUTO
Coding the JCL EXEC statement for producing configuration reports

PARM options to produce configuration reports. (The example later in this chapter describes the JOB and
ASSGN statements for an IOCP job.)
You must code the REPORT option to receive reports of an IOCDS.

PARM=’REPORT=A|Ax[,Ax...],LINECOUNT=n’
REPORT=
Specifies that IOCP reads one or more IOCDSs from the Support Element hard disk to produce
configuration reports. You can code more than one REPORT option. If you code only one option, you
can omit the parentheses.
Notes:
1. If you run z/VSE as a guest under z/VM, REPORT is not supported.
2. You cannot use ICPIOCP to produce reports for IOCDSs produced by the IYP, IZP, or IXP versions
of IOCP.
3. You can read an IOCDS only from an authorized logical partition (that is, I/O Configuration
The valid options for REPORT are Ax and A.
Ax Specifies the Ax IOCDS, where x is 0, 1, 2, or 3.
A Specifies all the Ax IOCDSs. See Appendix F, “Machine limits and rules,” on page 285 to
determine the Ax IOCDSs your CPC supports.
LINECOUNT=n
Specifies the number of lines that IOCP prints on a page of the IOCP output listing and configuration
reports, including the heading of the report. Code a decimal value in the range of 20-99.
If you do not use LINECOUNT, use the equivalent z/VSE command SET LINECT=n.
Conditions for coding the REPORT option

Code REPORT to apply to the following conditions:
v Use it only for IOCDSs written with ICPIOCP to avoid IOCP ending with an error.
v Use it in the absence of the WRTCDS and CHECKCPC options. You cannot read an input file and
generate an IOCDS (which includes reports of the generated IOCDS), and produce reports with the
REPORT option in the same job.
v Code several REPORT options; IOCP supports redundant or duplicate REPORT values. For example,
REPORT=(A0,A1,A2,A3) is the same as REPORT=A. Regardless of the order in which you specify the
options, IOCP produces reports in the following order: A0, A1, A2, A3.
JCL example of producing configuration reports

This example shows the JCL statements to produce IOCP reports of all the Ax IOCDSs read from the
Support Element hard disk. IOCP requires SYSLST for the output data set. It sends reports to SYSLST.
You do not need an ASSGN statement to access an IOCDS from the Support Element hard disk.
// JOB IOCP2
// ASSGN SYSLST,cuu
// EXEC PGM=ICPIOCP,SIZE=AUTO,PARM=’REPORT=A’

In the example:
JOB
SYSLST
EXEC
PGM=ICPIOCP
SIZE=AUTO
PARM='REPORT=A'
Specifies that IOCP reads all the Ax IOCDSs from the Support Element hard disk to produce
reports.
Return codes
When you run the z/VSE version of IOCP, it returns one of the following codes at the completion of the
job:
Table 13. Return codes for z/VSE version of IOCP
Return codes
0 0 IOCP has completed the requested function without error and might have sent
one or more informational messages. If so, the informational messages in the
output listings or the reports describe the specific conditions.
or more caution messages. The caution messages in the output listings or the
or more warning messages. The warning messages in the output listings or the
8 8 IOCP has encountered an error and has ended processing before completing
the requested function. The error message in the output listing describes the
specific condition.
16 10 IOCP has ended because it was unable to obtain the needed storage with the
GETVIS statement. To determine the partition size to use, see “IOCP storage
requirements” on page 131.
Methods of running the z/VSE version of IOCP

This topic suggests the steps that you can use to run the z/VSE version of IOCP to do the following
tasks:
v Define and check the I/O configuration data to prepare for a new CPC
v Redefine and replace an I/O configuration on an installed CPC.

The following example describes the steps that you can use to initially define and check the I/O
configuration data to prepare for the installation of a new CPC. The example suggests that you develop a
card-image input file (consisting of IOCP statements) and check the input file on an existing system. The
service representative uses this input file to generate the I/O configuration data during the installation of
the new CPC unless you are upgrading a CPC and were able to write an IOCDS in preparation of the
upgrade. Use this input deck to generate new I/O configuration data after changing your I/O
configuration.
1. Determine, from your physical plan, the I/O configuration requirements that you must meet to install
the CPC.
2. Prepare an input file (containing IOCP statements) that defines your I/O configuration. Using this file,
run IOCP as described in the following steps. (Appendix B, “Lists of statement input,” on page 241
shows a list of a typical IOCP input file containing IOCP statements.)
3. Run IOCP with JCL statements to validate the IOCP statements.
In the PARM parameter of the EXEC statement, use the WRTCDS=NO option. This option checks the
syntax of the IOCP statements and checks the configuration reports for the wanted configuration data.
5. Analyze the IOCP configuration reports (produced in Step 3) to ensure that you have defined I/O
configuration that you want. If necessary, revise the IOCP statements and rerun IOCP.
6. If you are upgrading to a CPC whose SYSTEM= value supports execution parameter
CHECKCPC=NO, write the IOCDS for the new CPC on your current CPC using CHECKCPC=NO.
See “Machine rules” on page 285 to determine which CPCs can write an IOCDS in preparation for an
upgrade and which CPCs can have ICPIOCP write an IOCDS to them. Inform the service
representative which IOCDS to use after the upgrade. This method is the preferred method for
writing the initial IOCDS when upgrading a CPC. Skip Step 7.
If you are upgrading to a CPC whose SYSTEM=value does not support execution parameter
CHECKCPC=NO or if this installation is a new installation, continue with Step 7.
7. To save the validated input file the service representative to use during installation of the CPC,
download the input file to diskette (if it is available) or USB flash memory drive. If the file is too large
to fit, compress it using a zip-compatible program. Source from a diskette must be in DOS file format
or must have been compressed. Source from a USB flash memory drive must be in VFAT or DOS file
format or have been compressed. Instead of a diskette or USB flash memory drive, an uncompressed
IOCP source file can be imported to the SE from an FTP location by the service representative. System
z9 and zSeries servers do not support USB flash memory drives for IOCDS source files.
Note: During the physical installation of the CPC, the service representative uses your validated input
file (with a stand-alone version of IOCP) to configure and test the I/O configuration for the new CPC.
See “Creating an IOCDS in a new CPC” on page 145 for an example using the stand-alone version of
IOCP.

The following example describes the steps you can use to replace the existing I/O configuration data on
a CPC. Use this procedure when the I/O configuration for the CPC changes and you must run IOCP.
1. Update your IOCP input file to reflect your new I/O configuration.
2. Run IOCP with JCL statements to validate the IOCP statements you updated in Step 1. In the PARM
parameter of the EXEC statement, use the WRTCDS=NO option.
4. Analyze the IOCP configuration reports (produced in Step 2) to ensure that you have defined the I/O
configuration that you want. If necessary, revise the IOCP statements and rerun IOCP (Step 2).

5. Run IOCP using JCL statements that specify a write option (WRTCDS=Ax). IOCP writes the specified
IOCDS to the Support Element hard disk. The system operator must respond YES to IOCP message
ICP050D before IOCP writes to the Support Element hard disk.
6. Update your z/VSE IPL procedure ($IPL...) to reflect the changes you made to the IOCP input file.
7. Perform a power-on reset of the CPC with the appropriate IOCDS specified as the source of your
configuration data. Click Disable Write Protection under Options on the Input/Output Configuration
window of the Support Element to remove write protection.
8. Test the new configuration.

Chapter 6. Running the stand-alone version of IOCP
The following information is an overview of the stand-alone version of the input/output configuration
program (IOCP). A separate document, Stand-Alone IOCP User's Guide, SB10-7173, provides detailed
instructions for using IOCP.
General information
The stand-alone version of IOCP is shipped with the CPC as part of the Licensed Machine Code. Run the
stand-alone version of IOCP to define the I/O configuration to the CPC before an operating system
version of IOCP is installed.
The stand-alone program uses the object-action user interface and is designed for use with a keyboard or
a mouse. If you are not familiar with using an object-oriented user interface, see “Related information” on
page xv for the Stand-Alone IOCP User's Guide appropriate to your system.
You can use the stand-alone version of IOCP to save the IOCDS source file to the hard disk of the
Support Element.
IOCDS source files on the SE hard disk can contain only 256,000 card records. Twenty additional records
are reserved for IOCP messages. If the selected IOCP function includes writing the IOCDS source file to
the SE and the file is too large, IOCP removes IOCP messages from the file to reduce its size until it is
small enough to be written.
The stand-alone version of IOCP must be run in a logical partition that was activated with a minimum of
256 MB (MB equals 1 048 576 bytes) of storage.
The stand-alone version of IOCP removes any comment card that begins with *ICP or *IYP. Stand-alone
IOCP uses *ICP comments for IOCP messages. When you generate an IOCDS, stand-alone IOCP removes
these IOCP messages so that the resultant IOCDS source file only contains IOCP messages from the
selected function.
The stand-alone version of IOCP does not have I/O recovery capability. I/O error conditions cause IOCP
to terminate.
Generating an input/output configuration data set

The CPC requires specific data about all attached control units and I/O devices. A text file must be
created to supply this information to the CPC. This file can be provided on diskette (if it is available) or
transferred from an FTP location to IOCP. If necessary, for a diskette, the file can be compressed using a
zip-compatible program. IOCP loads this data into the CPC and builds the IOCDS.
Note: The stand-alone version of IOCP is needed only:

v When you generate (Build action) an IOCDS during initial installation.
v If your operating system does not have a batch version of IOCP.
Creating an IOCDS source file for a new CPC

All logical partitions, channel paths, control units, and I/O devices must be defined in the IOCDS source
file to generate an IOCDS, which can then be used to perform power-on reset of the CPC.
Create an IOCDS source file as part of the installation procedure for a new CPC.

Use the text editor of your choice to create a file with the statements for the logical partitions, channel
paths (chpids), control units, and I/O devices.
v The format and syntax of the IOCP statements are explained in detail in Chapter 2, “IOCP statements
and functions,” on page 53.
v See Appendix D, “List of I/O devices and control units,” on page 245 for specific control unit and
device information.
v Examples of IOCP statements are in Stand-Alone IOCP User's Guide, SB10-7172, and Appendix E,
“Configuration examples,” on page 253.
To create IOCP input statements:

1. Create a drawing of the physical arrangement of control units and attached devices to show all
channel connections. You can use your own floor plan or some other source. It is much easier to be
certain you have included all of the I/O devices on your system in the IOCP statements when you
can see where all of the devices are. For more information about floor plans, see the Installation
Manual for Physical Planning for your CPC (for example, for System z10 EC, see the System z10
Enterprise Class Installation Manual for Physical Planning, SC28-6868).
2. Enter the statements. Use Chapter 2, “IOCP statements and functions,” on page 53 and Appendix D,
“List of I/O devices and control units,” on page 245 for specific information, and observe the
following tips:
v Enter the ID statement first. The first eight characters of the MSG1 parameter are the IOCDS name
displayed on the Hardware Management Console when the IOCDS is built. Make certain the
SYSTEM parameter matches your system.
v Enter the RESOURCE statement. Define each logical partition and logical channel subsystem.
v Enter the UUID statement if you want to enable UUID checking for any logical partitions.
v Enter the CHPID statements. Define each physical channel path installed on your system.
v Enter the CNTLUNIT statements. It is important to match the PROTOCL and UNITADD
parameters with the values set in each control unit by the system installer.
v Enter the IODEVICE statements
v Enter the FUNCTION statements.
After completing the input, save the edited file and exit the editor.
Use another CPC running an operating system version of IOCP (z/OS, z/VM, or z/VSE) to verify the
card-image input statements for this IOCDS source file. If you do not have access to an operating system
version of IOCP capable of verifying an IOCDS source file, ask your service representative to verify the
IOCDS source for you.
If you are upgrading to a CPC whose SYSTEM= value supports execution parameter CHECKCPC=NO,
write the IOCDS for the new CPC on your current CPC using CHECKCPC=NO. See “Machine rules” on
page 285 to determine which CPCs can write an IOCDS in preparation for an upgrade and which CPCs
can have an IOCDS written to them by ICP IOCP. Inform the service representative which IOCDS to use
after the upgrade.
If you are not upgrading or cannot write an IOCDS in preparation for the upgrade, use a utility available
to your personal computer and operating system to download the IOCDS source file to diskette (if it is
available) or USB flash memory drive. Source from a diskette must be in DOS file format or have been
compressed using a zip-compatible program. Source from a flash memory drive can be in VFAT or DOS
file formats or have been compressed. The IOCDS source file you create must be available to the new CPC
installer at the time the new CPC is installed. Instead of a diskette or USB flash memory drive, you can
import an uncompressed IOCP source file to the SE from an FTP location. System z9 and zSeries servers
do not support USB flash memory drives for IOCDS source files.

Starting the program
Start and control the stand-alone version of IOCP from the Hardware Management Console or SE
console. The CPC must be powered-on and a logical partition must be activated to build a new IOCDS.
Begin the I/O configuration procedure as follows:
v From the Support Element workplace window, click Task List from Views.
v Open CPC Configuration tasks.
v Open Groups from Views.
v Open Images from Groups Work Area
v Select an activated logical partition image icon in the Groups Work Area. The logical partition must be
activated with I/O Configuration Control Authority enabled through its activation profile.
v Drag the image icon (using the right mouse button) and drop it on the Input/Output (I/O)
Configuration task.
Input/Output configuration window

The Input/Output Configuration window of the Support Element is the starting point for all of the
stand-alone IOCP functions.
You can select a particular configuration to work with from this window or see the status of each
configuration. This window, found in the Stand-Alone IOCP User's Guide for your CPC, shows that there
are four IOCDSs identified as A0-A3, and these have names that you supply.
The fifth IOCDS, D0, is a diagnostic IOCDS. You cannot alter D0. It is intended for service personnel to
use, with the following possible exception. If you cannot perform a power-on reset and activate a logical
partition in ESA/390 mode using any of the A0-A3 IOCDSs, you can use D0 to perform the power-on
reset. I/O cannot be accessed after a D0 power-on reset.
The Stand-Alone IOCP User's Guide explains the Input/Output Configuration window of the Support
Element and the Options and View actions (see “Related information” on page xv). This window
provides access to all of the information about the IOCDSs currently available on the CPC. You can use
the functions under Options to change that information.
The Input/Output Configuration window of the Support Element displays the following information:
Data Set
Is a 2-character identifier for the IOCDS. There are five IOCDSs (A0, A1, A2, A3, and D0).
Name Is an 8-character name that identifies the IOCDS. The name comes from the first eight characters
that you have entered in the MSG1=keyword in the ID statement. The displayed name is saved from
the IOCDS when it is written to the Support Element.
Write Protected
Write-protection prevents batch IOCP or stand-alone IOCP from overwriting the IOCDS.
NO Indicates that the IOCDS is not write-protected.
YES
Indicates that the IOCDS is write-protected.
Date Is the month, day, and year when the IOCDS was built.
Time Is the hour and minute when the IOCDS was built. (Displayed in 24-hour format).
Data Set Status
Can be one of the following:
Valid Indicates the IOCDS is usable for power-on-reset purposes. Its content matches the source
only if the source status is Verified or Warnings.
Chapter 6. Running the stand-alone version of IOCP 141

Invalid
Specifies the IOCDS is not usable for power-on reset purposes. Either IOCP is currently
writing to the IOCDS or the IOCDS was written in preparation for a CPC upgrade and is
unusable until the CPC is upgraded to the type of CPC the IOCDS supports.
Active Specifies this IOCDS was used during the most recent power-on reset or was selected by
a dynamic activation from the SCP. Only one IOCDS can be active at a given time. The
active IOCDS is made write-protected by the Support Element during POR with the
IOCDS or after a dynamic activation by the SCP.
Source Status
Can be one of the following:
Verified
Indicates that the stand-alone IOCP Build was processed with the source for the selected
IOCDS and was successful with no warnings or errors.
Warnings
Indicates that the stand-alone IOCP Build was processed with the source for the selected
IOCDS and was successful with one or more warnings or cautions.
Errors Indicates that the stand-alone IOCP Build was processed with the source for the selected
IOCDS and was not successful. There was a terminal error.
Note: No IOCDS is written and the IOCDS remains as it was before the build.
Modified
Indicates that the source for the selected IOCDS has been changed by editing the file.
This value also indicates that the source might not match the current IOCDS.
Imported
Indicates that an import overlaid the source file.
Empty Indicates that the source for the selected IOCDS has been deleted or was erased when a
batch IOCP program wrote the IOCDS.
Version
Number that identifies which version of IOCP was used to create the IOCDS.
IOCP options
The list of actions is displayed when you click Options in the Input/Output Configuration window of
the Support Element.
You can select the action to be performed by clicking with the mouse or using the cursor movement and
Enter keys.
Some of the option actions (Enable/Disable Write Protection, Open Source File, Copy Configuration,
Import Source File, Export Source File, and Delete Source File) are performed using the Hardware
Management Console or Support Element console and do not require you to run stand-alone IOCP.
However, the Build Data Set, Disassemble Data Set, and Print Data Set Report actions automatically
start stand-alone IOCP and run it on the CPC. When one of these functions is selected, you receive a
warning stating that any CPC program or operating system running in the selected logical partition will
shut down for stand-alone IOCP to perform these functions.
Following is a brief overview of each Options action. For detailed descriptions of these actions, see
“Related information” on page xv for the Stand-Alone IOCP User's Guide appropriate for your system.

Overview of options
Enable/Disable Write Protection
Enable protects the IOCDS and the IOCP source file on the hard disk from being overwritten. Disable
resets the overwrite protection.
Copy Configuration
Copies one IOCDS to another. You might use this option to duplicate an existing I/O configuration so
you can test some changes on the copy. Both the IOCDS source file and the IOCDS are copied.
Open Source File

Calls an editor to modify the IOCDS source file card-image data for the selected configuration or to create
new card-image data. (See “Creating an IOCDS in a new CPC” on page 145.)
Import Source File

Transfers the card image input to the CPC Support Element hard disk from any of the following, if
available:
v A USB flash memory drive in the Hardware Management Console
v An FTP location
v A diskette in the Hardware Management Console's 3.5-inch A-drive
The IOCDS source file can then be edited and used to build an IOCDS. If you are importing from
diskette, IOCDS source files must be in DOS file format or have been compressed using a zip-compatible
program. The source file name must have a maximum of 8 characters, and the file extension must have a
maximum of 3 characters. If you are importing from a USB flash memory drive, IOCDS source files can
be in VFAT or DOS file format or have been compressed.
Note: USB flash memory drives are supported only for CPCs with fixed HSA. See Table 16 on page 285
to determine if your CPC has fixed HSA.
Export Source File

Transfers the card-image IOCDS source file from the CPC Support Element hard disk to any of the
following, if available:
v An FTP location
v A diskette in the A-drive of the Hardware Management Console
Delete Source File

Note: Before using Delete Source File, make a backup of the IOCDS source file. (Use the Export action
to make the backup file.)
Erases the IOCDS source file and replaces it with a minimum default template file. Only the source file is
deleted. The IOCDS file is not deleted.
Note: Use this action with extreme care, because the IOCDS source file for the selected IOCDS is erased.
Build Data Set

Does the following:
v Processes the IOCDS source file for the selected IOCDS.
v Checks the syntax and validates the configuration.
v Generates an IOCDS and, if there are no errors, writes the IOCDS to the CPC Support Element hard
disk.

v Writes the IOCDS source file to the CPC Support Element hard disk with any IOCP messages
embedded in the file.
Disassemble Data Set

Generates card-image statements (IOCDS source file) from an IOCDS on the CPC Support Element hard
disk and writes an IOCDS source file. The new IOCDS source file contains the full I/O configuration and
is the logical equivalent of the original IOCDS source file.
Print Data Set Report

Prints or writes to tape a formatted report of the IOCDS.
Cancel
Leaves the I/O Configuration Program.
IOCP view actions

Use the View actions to see what is written in the selected IOCDS. The information displayed using View
is read directly from the IOCDS, not the IOCDS source file. Because both control unit and device
information are displayed on a per-channel basis, you must select the channel (CHPID) to be displayed.
Methods of running the stand-alone version of IOCP

Following are descriptions of various ways to run the stand-alone version of IOCP.
Before system installation, the service representative must find out if the IOCDS source file has been
previously prepared for use during installation. If you are upgrading a CPC and an IOCDS was written
in preparation of the upgrade, use that IOCDS. Otherwise, a new IOCDS must be written.
If the IOCDS source file has been previously prepared, then it must be imported to the CPC Support
Element. The IOCDS source file can be on any of the following, if supported:
v A location that can be accessed through FTP (uncompressed text file)
v A 3.5-inch diskette (DOS format or compressed file)
If no IOCDS or IOCDS source file is available, then it must be created at the time of installation. Use the
editor supplied with the stand-alone I/O Configuration program to create the IOCDS source file.
If a default or old IOCDS source file exists, you can use the Delete source file action to clear the file or
use the editor actions to Mark and Delete the default or existing statements in the selected IOCDS source
file rather than creating a new one.
Run the stand-alone version of IOCP to:

v Create a new IOCDS during the installation of a new CPC if the IOCDS source file is not available.
v Update an IOCDS on an existing CPC, using the Open Source File action. In some cases where the
IOCDS source file was not saved or if you are unsure of the IOCDS source file, you can select the
Disassemble Data Set action to get an equivalent IOCDS source file from an existing IOCDS.
v Replace an IOCDS (using Import and Build) in a CPC if the IOCDS source file is available on a
diskette.
Starter IOCDS information for the stand-alone version of IOCP

Four starter IOCDSs are shipped with the machine for your use, and an additional diagnostic IOCDS is
intended for service representatives to use. The starter IOCDSs are provided as examples only and are

not intended to match your actual I/O configuration. You can view the contents of the starter IOCDSs by
using any action under View or the Open Source File action under Options on the Input/Output
Configuration window of the Support Element.
Creating an IOCDS in a new CPC

All IOCDS source files consist of RESOURCE, CHPID, CNTLUNIT, IODEVICE, and FUNCTION
statements. All logical partitions, paths, control units, I/O devices, and functions must be defined in the
IOCDS source file to generate an IOCDS, which can then be used to power-on reset the CPC.
Perform the steps that create an IOCP source file as part of the installation procedure for a new CPC.
The following example assumes that the IOCDS source file must be entered from scratch. All of the
logical partitions, I/O channel paths, control units, and installed device unit addresses must be defined
and available to the service representative. When the new IOCDS source file is created, it defines the
initial I/O configuration for the new CPC. Comments are saved.
To create a new IOCDS for a new CPC complex:

1. Perform a power-on reset of the system with the D0 IOCDS and activate a logical partition
2. Select the I/O Configuration window. See “Starting the program” on page 141.
3. Select the IOCDS to create.
4. On the Input/Output Configuration window of the Support Element, click Options and then Delete
Source File to remove the existing IOCDS source file for the selected configuration. (If there is a
default or example IOCDS source file, the service representative must check with the IOCDS source
file owner to determine whether to save this file. (The service representative can use the Export
Source File action to move the IOCDS source file to a diskette (if it is available)).
5. Enter the statements for the logical partitions, channel paths (chpids), control units, and I/O devices.
v The format and syntax of the IOCP statements are explained in detail in Chapter 2, “IOCP
statements and functions,” on page 53.
v See Appendix D, “List of I/O devices and control units,” on page 245 for specific control unit and
device information.
v Examples of IOCP statements are in Appendix E, “Configuration examples,” on page 253.
To create an IOCDS source file:
a. Use a drawing of the physical arrangement of control units and attached devices to show all
channel connections. This drawing might be from your own floor plan or some other source. It is
much easier to be certain you have included all of the I/O devices on your system in the IOCP
statements when you can see where all of the devices are. For more information about floor plans,
see the Installation Manual for Physical Planning for your CPC (for example, for System z10 EC, see
the System z10 Enterprise Class Installation Manual for Physical Planning, SC28-6868).
b. Enter the statements. Use Chapter 2, “IOCP statements and functions,” on page 53 and
Appendix D, “List of I/O devices and control units,” on page 245 for specific information, and
observe the following tips:
v Enter the ID statement first. The first eight characters of the MSG1 parameter are the IOCDS
name displayed on the Hardware Management Console when the IOCDS is built. Make certain
the SYSTEM parameter matches your system.
v Enter the RESOURCE statement. Define all logical partitions and logical channel subsystems.
v Enter the UUID statement if you want to enable UUID checking for any logical partitions.
v Enter the CHPID statements. Define each physical channel path installed on your system.
v Enter the CNTLUNIT statements. It is important to match the PROTOCL and UNITADD
parameters with the values set in each control unit by the system installer.
v Enter the IODEVICE statements.
v Enter the FUNCTION statements.

After completing the input, save the edited IOCDS source file and exit the editor.
6. Select the Build Data Set action to generate the system IOCDS. When successfully completed, the
IOCDS is written to the CPC Support Element hard disk. If the build is not successful, IOCP error and
warning messages are inserted in the card-image statements, and a copy of the IOCDS source file is
saved on the CPC support Element hard disk. Return to the editor to correct the errors.
7. Exit the Input/Output Configuration window of the Support Element.
8. Perform a power-on reset of the system with the IOCDS written in Step 6. Select the newly created
IOCDS and the appropriate system mode in the Power-on Reset window.
Updating an IOCDS using the card-image input editor

This procedure describes how to update the I/O configuration data in an IOCDS for an installed CPC.
To update the I/O configuration for a CPC:

1. Ensure that the CPC is in a power-on-reset state with an appropriate existing IOCDS.
2. Select the I/O Configuration task. See “Starting the program” on page 141.
3. Select the IOCDS to update.
4. Check to see if the IOCDS source file matches the contents of the IOCDS. (The source status is
Verified or Warnings, and the IOCDS status is Valid when the IOCDS was built from the IOCDS
source file). If the status for the IOCDS source file and the IOCDS do not match, then the IOCDS
source file for the selected IOCDS might be different than the IOCDS source file used to create the
IOCDS. The IOCDS source file might have been imported from diskette (if it is available) or it might
have been modified using the editor. You can create a reconstructed IOCDS source file by selecting
the Disassemble Data Set action. The Disassemble Data Set action reads the selected IOCDS and
creates an equivalent to the IOCDS source file used to create the IOCDS.
5. If you are adding (or deleting) I/O from your configuration, update the IOCDS source file (Open
Source File) to reflect any new I/O configured in the CPC.
6. Prepare the system line printer or channel-attached tape drive if you want printed output. The
printer must be initialized (the print/forms buffer must be loaded).
7. Select the Build Data Set action under Options on the Input/Output Configuration window of the
Support Element.
a. Select the appropriate parameters and press Enter.
During the build, IOCP reads the input statements, saves them as card-image data in storage,
checks the syntax, validates it, and generates an IOCDS. It produces configuration reports of the
IOCDS generated. (See Chapter 2, “IOCP statements and functions,” on page 53 for more details.)
IOCP displays messages to give the status of the IOCDS build.
b. An error message is displayed if IOCP has detected errors in the IOCDS source file. To correct
the errors, on the Input/Output Configuration window of the Support Element after the
completion message, click Options and then Open Source File.
Errors IOCP detects are displayed in the editor. Errors are displayed in the form of IOCP
comments that begin with *ICP in column 1, followed by the error message number and the
message text. See Chapter 8, “IOCP messages,” on page 171 for an explanation of the error
messages. Correct the errors in the IOCDS source file. It is not necessary to remove the error
messages from the file. IOCP removes the messages when no errors are detected on subsequent
builds.
8. View the generated IOCDS content using the View action (Control Unit Information and Device
Information windows) or analyze the configuration reports produced in Step 7a to ensure that the
I/O configuration you want has been generated. If the IOCDS source file must be changed, use the
editor to modify the IOCDS source file on the Hardware Management Console and return to the
Build Data Set process (Step 7).
9. If the input file is correct and the build is complete, the IOCDS is on the CPC Support Element hard
disk.

10. Export the IOCDS source file to provide a backup if editing was done on the Hardware Management
Console or SE console.
11. Exit IOCP by clicking the Cancel action on any IOCP window.
The I/O configuration data in the new Ax IOCDS now contains your new configuration. Selecting a
power-on reset, with Ax as the IOCDS parameter, activates the new configuration and makes the Ax
IOCDS the configuration that the I/O subsystem uses to control I/O requests. Select the IOCDS and
the system mode on the Power-on Reset window.
12. Test the new configuration following the procedures established for your installation.
Notes:
1. If you change the hardware configuration with the stand-alone version of IOCP and you use a z/OS
system or if you use a VM system on which you have chosen to use HCM and HCD, make the same
changes using HCM or HCD and rewrite the IOCDS before performing the power-on reset. This
action ensures consistency between the software and hardware configurations and makes the CPC
capable of dynamic I/O configuration.
2. If you change the hardware configuration with the stand-alone version of IOCP and you use a VM
system but do not use HCM and HCD, make any appropriate changes to the VM system before
performing the power-on reset. If VM cannot sense all the device changes in the hardware
configuration during initial program load (IPL), modify the real I/O configuration file and regenerate
your operating system, or modify the system configuration file and re-IPL your operating system.
These actions ensure consistency between the software and hardware configurations. (The real I/O
configuration file is HCPRIO. The system configuration file is SYSTEM CONFIG.)
For a CPC capable of dynamic I/O configuration, you must rewrite the IOCDS from VM using the
DYN option and perform a power-on reset.
3. For a VSE system, if you change the hardware configuration with the stand-alone version of IOCP,
update your software configuration to reflect the changes (modify the ASI Procedure file). Doing so
ensures consistency between the software and hardware configurations.
Description of a stand-alone build of an IOCDS

IOCP performs the following steps when you specify that you want to generate or build an IOCDS using
the Build data set action.
1. The CPC Support Element resets the logical partition.
2. The CPC Support Element loads stand-alone IOCP into central processor storage for the targeted
logical partition and starts the program.
3. IOCP reads the IOCDS source file from the CPC Support Element hard disk.
Note: IOCP code Steps 4-5 are repeated for each input statement.
4. IOCP checks the syntax of each statement for correct format. If a syntax error is found (such as a
missing parenthesis), IOCP inserts an error message in the IOCDS source file following the statement
that contains the error. IOCP stops checking the syntax of a statement when it encounters the first
syntax error on that statement. Therefore, if there are two or more syntax errors on a single
statement, the second and following errors are found on subsequent runs of IOCP.
If IOCP finds an error in Step 4, it performs Steps 4 and Step 5 for the remainder of the statements
and then continues with Step 9 on page 148.
5. IOCP checks the validity of each statement. For example, IOCP checks that a CNTLUNIT statement
has specified a channel path that has been previously defined in a CHPID statement. If an error is
found, IOCP inserts an error message in the IOCDS source file following the statement that contains
the error.
If IOCP finds an error in Step 5, IOCP returns to Step 4, and performs only Steps 4 and 5 for the
remainder of the statements and then continues with Step 9 on page 148.

B20 • B22
6. After all of the input statements have been processed, IOCP generates the IOCDS and makes
additional checks. For example, IOCP checks that the maximum number of control unit headers is
not exceeded. If IOCP finds an error, IOCP issues an error message to the Hardware Management
Console.
7. IOCP produces IOCP configuration reports of the IOCDS.
8. IOCP writes the IOCDS to the CPC Support Element hard disk if no errors were found.
9. IOCP writes the IOCDS source file to the CPC Support Element hard disk. If the build was
successful, then the IOCDS source file has any previous error messages removed. If errors have been
detected, they are embedded in the IOCDS source file.
10. The Hardware Management Console or SE console ends the IOCP program. (Starting stand-alone
IOCP after this point causes IOCP to time out and to put the logical partition in a B20 hard wait
state.)
IOCP wait state codes
Enabled wait states

When you run the stand-alone version of IOCP, the central processor enters many enabled wait states
(PSW bit 7 is on). These are normal conditions while IOCP waits for the Support Element to respond to
IOCP requests such as a read or write of an IOCDS. You can either wait for the Support Element to finish
or click Cancel and restart IOCP at a later time.
Disabled wait states

When you run the stand-alone version of IOCP, IOCP sends the following disabled wait state codes when
the indicated conditions occur.
Note: z/OS, z/VM, and z/VSE versions of IOCP do not send wait state codes.
B20
Explanation: The operator who is running the stand-alone version of IOCP has started IOCP processing after the
IOCP function has completed, or IOCP was unable to read from the the selected function to perform.
Operator response: If IOCP completed the selected function, no action is required. Otherwise, notify your system
programmer.
Problem determination: If IOCP did not complete the selected function, save all associated output. Contact the
service representative.
B21
Explanation: A program-check interruption has occurred during processing of the stand-alone version of IOCP.
Operator response: Notify your system programmer.
Problem determination: Save all associated output. Call your service representative.
B22
Explanation: A machine-check interruption has occurred during processing of the stand-alone version of IOCP.

B23 • B24
B23
Explanation: An uncorrectable I/O error has occurred during processing of the stand-alone version of IOCP.
B24
Explanation: During processing of the stand-alone version of IOCP, IOCP has encountered an error in the and
cannot communicate with the person operating IOCP.

Chapter 7. IOCP configuration reports
This chapter describes the IOCP configuration reports. Use the reports to check the I/O configuration
data in an IOCDS. Following is a brief description of each configuration report with references to report
examples. If you have any comments about the IOCP reports, see “How to send your comments” on
page xx.
Configuration report headings

Seven heading lines are present for each configuration report. Line 1 shows:
v Time and date IOCP produced the reports.
v IOCP version and release number of the IOCP release producing the reports Type of configuration
report (IOCDS TOTALS, PARTITION, FID SUMMARY, PCHID SUMMARY, CHPID SUMMARY, I/O
DEVICE, or CHANNEL PATH).
v Report page number.
Line 2 shows:
v Version and release of IOCP that wrote the IOCDS when the source of the report is an IOCDS from the
Support Element hard disk.
v Blank line if the source of the report is card input; the version and release are not shown.
Line 3 shows:
v Source of the report, which can be card input, an IOCDS from the Support Element hard disk, or
screen input (stand-alone editor).
v Whether the IOCDS was ever modified by the stand-alone editor (YES or NO).
v CHECKCPC=NO if you specified this keyword in the PARM parameter of the EXEC statement or CMS
option NOCHKCPC to write an IOCDS in preparation of an upgrade.
Line 4 (ID1=) shows:

v Any identification information that you specified with the MSG1 parameter in the ID statement. (The
first eight characters of ID1= are the same as the customer name of the IOCDS that is on the
Input/Output Configuration window of the Support Element.)
v The time and date IOCP last wrote the IOCDS for a report source that is an IOCDS from the Support
Element hard disk.
Line 5 (ID2=) shows:

v Any additional identification information that you specified with the MSG2 parameter of the ID
statement.
v The Support Element time stamp, when the report source is an IOCDS from the SE hard disk. (The SE
time stamp also is present on the Input/Output Configuration window of the Support Element.)
Line 6 (SYSTEM=) shows:

v The SYSTEM keyword value used by IOCP to determine which machine limits and rules to enforce.
v Whether IOCP performs a deck verification, attempts to write an IOCDS, or reports on an IOCDS from
the Support Element hard disk.
v A warning if you are reporting on an IOCDS that was written to your CPC when the CPC was
represented by a different SYSTEM keyword value than the one that currently represents it. For

example, you have upgraded your CPC and a different SYSTEM keyword value now represents it. Do
not use this IOCDS for power-on reset unless the I/O configuration for your current CPC is identical
to the configuration in the IOCDS.
Line 7 (DYNAMIC INFO:) shows:

v Any dynamic I/O configuration information associated with the IOCP source input or an IOCDS that
supports dynamic I/O configuration.
v IOCDS IS NOT DYNAMIC CAPABLE if IOCDS does not support dynamic I/O configuration.
v The LSYSTEM keyword value specified on the ID statement or <NONE> if the keyword was not
specified and the CPC supports CIB, CL5, or CS5 channel paths.
v For the IOCDS TOTALS report, the DYNAMIC INFO line is reported on line 8 and line 7 is blank.
IOCDS totals report

The IOCDS Totals Report lists the configuration totals in the IOCDS. Three groupings of IOCDS totals are
shown.
v Channel paths, FIDs, and physical control units
v CHPIDs and control unit headers for each CSS ID
v Devices for each subchannel set in each CSS ID.
“Sample IOCDS totals report” on page 153 shows an example of the IOCDS Totals Report.
Note: Prior to the introduction of the report field for the total number of FIDs, IOCP included the total
number of PCHIDs and VCHIDs for FIDs with the total number of channel paths.
For CPCs that have a fixed storage size for the Hardware System Area (HSA), the MAXDEV keyword is
not used and the maximum resources supported by your CPC are allowed.
For CPCs that do not have fixed HSA, the following paragraphs apply.
If an IOCDS supports dynamic I/O configuration, the MAXDEV keyword specified on the RESOURCE
statement is shown.
Each grouping of IOCDS totals includes a column for MAX ALLOWED, which is the maximum number
allowed for your CPC. For channel paths, the report column is always N/A. For devices for an IOCDS
that supports dynamic I/O configuration, the maximum number allowed is based on the MAXDEV
keyword values specified on the RESOURCE statement.
Each grouping of IOCDS totals includes a column for DYNAMIC SPARES. If an IOCDS supports
dynamic I/O configuration, this column shows the number of extra units that can be dynamically added
to the configuration before exceeding the maximum number allowed. For channel paths, the report
column is always N/A. If an IOCDS does not support dynamic I/O configuration, the report column is
N/A for all totals.
IOCP lists informational message ICP069I at the end of this report when IOCP selects a nonzero default
MAXDEV value.
A legend at the end of the report indicates which MAX ALLOWED column entries for devices are the
result of the default MAXDEV value being selected by IOCP or a value of MIN or MAX being specified.

Sample IOCDS totals report
TIME 15.11 DATE 17.094 REPORT BY ICP IOCP VERSION 5 RELEASE 4.0 IOCDS TOTALS REPORT PAGE 003
REPORT SOURCE IS CARD INPUT, MODIFIED BY SCREEN INPUT=NO

ID1=SYSTEM1
ID2=HCDI.IODF02.PROD - 2017-04-03 13:54
SYSTEM=(3906,1) USED BY IOCP FOR DECK VERIFICATION
DYNAMIC INFO: PROC1 04-03-17 13:54:42 HCDI IODF13 LSYSTEM=CPC1
MAXDEV <NOT USED BY PROCESSOR>
TOTALS FOR CHANNEL PATHS, FIDS, AND PHYSICAL CONTROL UNITS:

___________________________________________________________
IOCDS MAX DYNAMIC

TOTAL ALLOWED SPARES
_____ _______ _______
CHANNEL PATHS ---------- 23 N/A N/A

FIDS ------------------- 7 18432 18425
PHYSICAL CONTROL UNITS - 14 8192 8178
TOTALS FOR CHPIDS AND CONTROL UNIT HEADERS:

___________________________________________
CONTROL UNIT HEADERS

IOCDS MAX DYNAMIC
CSS CHPIDS TOTAL ALLOWED SPARES
___ ______ _______________________
0 15 7 4096 4089
1 9 10 4096 4086
2 0 0 4096 4096
3 4 3 4096 4093
4 0 0 4096 4096
5 0 0 4096 4096
TOTALS FOR DEVICES:

___________________
SUBCHANNEL SET 0 SUBCHANNEL SET 1 SUBCHANNEL SET 2 SUBCHANNEL SET 3

IOCDS MAX DYNAMIC IOCDS MAX DYNAMIC IOCDS MAX DYNAMIC IOCDS MAX DYNAMIC
CSS TOTAL ALLOWED SPARES TOTAL ALLOWED SPARES TOTAL ALLOWED SPARES TOTAL ALLOWED SPARES
___ ____________________________ ____________________________ ____________________________ ____________________________
0 214 65280 (MAX) 65066 64 65535 (MAX) 65471 0 65535 (MAX) 65535 0 65535 (MAX) 65535
1 241 65280 (MAX) 65039 0 65535 (MAX) 65535 0 65535 (MAX) 65535 0 65535 (MAX) 65535
2 0 65280 (MAX) 65280 0 65535 (MAX) 65535 0 65535 (MAX) 65535 0 65535 (MAX) 65535
3 224 65280 (MAX) 65056 64 65535 (MAX) 65471 0 65535 (MAX) 65535 0 65535 (MAX) 65535
4 0 65280 (MAX) 65280 0 65535 (MAX) 65535 0 65535 (MAX) 65535 0 65535 (MAX) 65535
5 0 65280 (MAX) 65280 0 65535 (MAX) 65535 0 65535 (MAX) 65535 0 65535 (MAX) 65535
LEGEND FOR MAX ALLOWED DEVICES FIELD:

_____________________________________
(***) - INDICATES THAT MAXIMUM NUMBER OF DEVICES ALLOWED VALUE WAS NOT SPECIFIED ON MAXDEV KEYWORD AND IS A DEFAULT
(MIN) - INDICATES THAT MAXDEV KEYWORD VALUE WAS SPECIFIED AS ’MIN’ AND THEREFORE THE MAXIMUM NUMBER OF DEVICES ALLOWED
EQUALS THE NUMBER OF DEFINED DEVICES (IOCDS TOTAL)
(MAX) - INDICATES THAT MAXDEV KEYWORD VALUE WAS SPECIFIED AS ’MAX’ OR CPC HAS FIXED HSA AND THEREFORE THE MAXIMUM NUMBER
OF DEVICES ALLOWED EQUALS THE MAXIMUM NUMBER SUPPORTED BY THE CPC
Figure 20. Sample IOCDS totals report
Partition report
The Partition Report lists the MIF image IDs (MIF IDs) defined in each CSS in the IOCDS and the logical
partition names to which they correspond. Undefined MIF image IDs are not shown. Each partition also
reports whether or not it has UUID checking enabled. The report also lists all I/O cluster names
(IOCLUSTER keyword) and the CSSs to which they have access. “Sample partition report” on page 154
shows an example of a Partition Report.
Chapter 7. IOCP configuration reports 153

Sample partition report
TIME 15.11 DATE 17.094 REPORT BY ICP IOCP VERSION 5 RELEASE 4.0 PARTITION REPORT PAGE 004

ID1=SYSTEM1
ID2=HCDI.IODF02.PROD - 2017-04-03 13:54
MIF PARTITION
ID NAME UUID
___ _________ ____
CSS 0:
1 BBBBB NO
2 LP0A NO
3 AAAAAAAA NO
4 LP0B NO
5 LP0C NO
6 LP0D NO
7 LP01 YES
8 LP02 NO
9 LP03 NO
A LP04 NO
B LP05 NO
C LP06 NO
D LP07 NO
E LP08 NO
F LP09 NO
CSS 1:
1 SYS1 YES
2 LP12 NO
3 *
4 CCCCCCC NO
5 LP15 YES
6 *
7 *
8 LP18 NO
9 *
A *
B CF1 NO
C *
D *
E *
F LP1F NO
CSS 2: (Pruned)
CSS 3:
1 VM1 NO
2 VSE1 NO
3 ZOS1 NO
4 *
5 *
6 *
7 *
8 *
9 *
A *
B *
C *
D *
E *
F *
CSS 4: (Pruned)
CSS 5:
1 ZOS5 NO
2 *
3 *
4 *
5 *
6 *
7 *
8 LP58 NO
9 *
A *
I/O CLUSTER CSS NUMBERS

NAME |0|1|2|3|4|5|
___________ _____________
SYSPLEX1 | |1| | | | |
Figure 21. Sample partition report

FID summary report
The FID Summary Report lists the function IDs (FIDs) defined in the IOCDS. For each FID, the report
shows the following:
v FID number
v PCHID number, PCHID and PORT numbers, or VCHID number the FID is associated with. VCHID
numbers are indicated by '(V)'.
v VF number
v FID type
v UID number
v PNETID values
v Partitions that can use the FID. The partitions are identified by their CSS and MIF image ID number.
“Sample FID summary report” on page 156 shows an example of the FID Summary Report.
If the VF or UID keywords are not specified, dashes are reported. If the PNETID keyword was not
specified or a PNETID keyword value was skipped, dashes are reported. However, if the UID or PNETID
keyword was not specified but IOCP provided a default value, the default value is reported.
If UUID checking is enabled for a partition that can access a FID, the UID value reported is appended
with '*'.
If a FID type does not require the PORT keyword and multiple PNETID values are specified for the
adapter's ports, the PNETID values are reported on successive lines. The first line is associated with the
PNETID for port 1, the second line is associated with the PNETID for port 2, and so on.
The MIF IMAGE ID NUMBERS report column shows the MIF image IDs of logical partitions that the FID
can access in the associated CSS. The CSS number that precedes a MIF image ID field is the CSS for
which that field applies.
v If a partition is in the access list of the FID, the MIF image ID number column contains an A for the
corresponding partition.
v For each partition in the candidate list but not the access list of the FID, the MIF image ID number
column contains a C for the corresponding partition.
v If a partition is in the access list of the FID and the partition has UUID checking enabled, the MIF
image ID number column contains an '&' for the corresponding partition.
v For each partition in the candidate list but not the access list of the FID and the partition has UUID
checking enabled, the MIF image ID number column contains a U for the corresponding partition.
v For all partitions not in the access or candidate lists of the FID, the corresponding MIF image ID
number columns are left blank.
A legend is located at the end of the FID Summary Report that describes the entries in the MIF IMAGE
ID NUMBERS report column.
At the end of the FID Summary Report, the totals are listed for each FID type your CPC supports.

Sample FID summary report
TIME 15.11 DATE 17.094 REPORT BY ICP IOCP VERSION 5 RELEASE 4.0 FID SUMMARY REPORT PAGE 006

ID1=SYSTEM1
ID2=HCDI.IODF02.PROD - 2017-04-03 13:54
MIF IMAGE ID NUMBERS MIF IMAGE ID NUMBERS

FID PCHID VF TYPE UID PNETID CSS |1|2|3|4|5|6|7|8|9|A|B|C|D|E|F| CSS |1|2|3|4|5|6|7|8|9|A|B|C|D|E|F|
_____ _____ ___ ____ ____ ________________ ___ _______________________________ ___ _______________________________
0001 200 1 ROCE 0012* NET1 0 | |C| | |A|C|U| |C| | | | | | | 1 | |C| | | | | |C| | | | | | | |
NETWORK2 2 | | | | | | | | | | | | | | | | 3 |C| |C| | | | | | | | | | | | |
4 | | | | | | | | | | | | | | | | 5 | | | | | | | |C| | | | | | | |
0180 204 7 ZEDC 0180* ---- 0 |C|C|C|C|C|C|U|C|C|C|C|C|C|C|C| 1 |U|A| |C|U| | |C| | |C| | | |C|
2 | | | | | | | | | | | | | | | | 3 |C|C|C| | | | | | | | | | | | |
4 | | | | | | | | | | | | | | | | 5 |C| | | | | | |C| | | | | | | |
1300 210-1 5 ROC2 1300* ---- 0 |C|C|C|C|C|C|U|C|C|C|C|C|C|C|C| 1 |U|A| |C|U| | |C| | |C| | | |C|
2 | | | | | | | | | | | | | | | | 3 |C|C|C| | | | | | | | | | | | |
4 | | | | | | | | | | | | | | | | 5 |C| | | | | | |C| | | | | | | |
1301 210-2 5 ROC2 F123* NET1 0 |C|C|C|C|C|C|U|C|C|C|C|C|C|C|C| 1 |&|C| |C|U| | |C| | |C| | | |C|
2 | | | | | | | | | | | | | | | | 3 |C|C|C| | | | | | | | | | | | |
4 | | | | | | | | | | | | | | | | 5 |C| | | | | | |C| | | | | | | |
2100 208 -- RCE ---- ---- 0 | | | | | | | | | | | | | | | | 1 | | | | | | | | | | | | | | | |
2 | | | | | | | | | | | | | | | | 3 | | |A| | | | | | | | | | | | |
4 | | | | | | | | | | | | | | | | 5 | | | | | | | | | | | | | | | |
3500 218-2 127 HYL 0123 ---- 0 |A| | |C|C| | | | | | | | | | | 1 | | | | | | | | | | | | | | | |
2 | | | | | | | | | | | | | | | | 3 |C|C|C| | | | | | | | | | | | |
4 | | | | | | | | | | | | | | | | 5 |C| | | | | | | | | | | | | | |
4000 7E5(V) 2 ISM ---- ABCDNETWORK 0 | |C| | | | | |A| | | | | | |C| 1 | | | | | | | | | | | | | | | |
2 | | | | | | | | | | | | | | | | 3 | | | | | | | | | | | | | | | |
4 | | | | | | | | | | | | | | | | 5 | | | | | | | | | | | | | | | |
LEGEND FOR MIF IMAGE ID NUMBERS FIELD:

______________________________________
A - PARTITION IN CHPID’S ACCESS AND CANDIDATE LISTS

C - PARTITION IN CHPID’S CANDIDATE LIST ONLY
& - PARTITION IN CHPID’S ACCESS AND CANDIDATE LISTS AND HAS UUID CHECKING ENABLED
U - PARTITION IN CHPID’S CANDIDATE LIST ONLY AND HAS UUID CHECKING ENABLED
BLANK - PARTITION NOT IN CHPID’S ACCESS OR CANDIDATE LIST
TOTALS FOR FID TYPES:

_____________________
HYL ISM RCE ROCE ROC2 ZEDC

___ ___ ___ ____ ____ ____
1 1 1 1 2 1
Figure 22. Sample FID summary report
PCHID summary report

The PCHID Summary Report lists the PCHIDs defined in the IOCDS and the channel paths that cannot
have PCHIDs. The report does not include channel paths that require a PCHID when an IOCDS is
written and for which no PCHID has been defined. It includes channel paths that cannot have a PCHID
at the bottom of the report. The report also includes PCHIDs or VCHIDs associated with a FID. For each
channel path or FID, the report shows the following:
v PCHID number, PCHID and PORT numbers, VCHID number, or AID and PORT numbers associated
with the channel path or FID. VCHID numbers are indicated by '(V)'. If the MIXTYPE keyword was
specified for the PCHID, this is indicated by '(M)'.
v CHPID number of the channel path
v Type of channel path or FID
v Switch number associated with the channel path
v CSS IDs to which the channel path has access
v PNETID values for the channel path or FID
v Any configuration discrepancy or comment for the channel path or FID.
Figure 23 on page 159 shows an example of the PCHID Summary Report.
If the PCHID, VCHID, and AID keywords are not allowed for the channel path type, the report column
contains N/A. If the report was produced for a deck verification and channel paths that require the PCHID
keyword did not specify PCHID, the channel paths are excluded from the report and a warning line is
printed at the end of the report. PCHIDs are separated by a blank line on a modulo-16 basis which on
some CPCs indicates that the PCHIDs are associated with different card slots..

If a FID is being reported, the CHPID and SWITCH report columns contain N/A.
If the SWITCH keyword is not allowed for the channel path type, the report column contains N/A. If the
SWITCH keyword is allowed but not specified for a channel path, then the column contains NONE.
If a CHPID or FID type supports multiple PNETID values, the PNETID values are reported on successive
lines. The first line is associated with the PNETID for port 1, the second line is associated with the
PNETID for port 2, and so on.
The CHANNEL CONFIGURATION DISCREPANCY OR COMMENT report column indicates whether a

channel path has a configuration discrepancy. The report column also displays a comment for certain
channel path types when the channel path does not have a configuration discrepancy. NO indicates no
configuration discrepancy exists. YES indicates that a configuration discrepancy exists and is accompanied
by a description of the discrepancy. More than one configuration discrepancy can exist for a channel path
but only one discrepancy is reported. The report column displays a comment for FID types.
Channel configuration discrepancies are conditions that should not exist on your CPC. Because IOCP
processes channel paths in ascending PCHID number order, it treats the lower numbered PCHIDs as
accurate and reports discrepancies against the higher numbered PCHIDs. An IOCDS is produced but
IOCP reports the discrepancies so that they can be corrected. For an explanation of the channel
configuration rules for your CPC, see Appendix F, “Machine limits and rules,” on page 285.
If no configuration discrepancy exists, a comment is displayed for certain channel path types and all FID
types. The following comments are displayed:
v Channel path types that support a maximum number of valid subchannels (for example, FCP and
OSD) display the current number of valid subchannels defined on the channel path. For OSN channel
paths, the OSN and 3745 subchannel totals are both displayed. For IQD channel paths, the number of
subchannels defined on the channel path and the total number of subchannels for all IQD channel
paths in the configuration are both displayed. For more information about the maximum subchannels
supported for a channel path type, see Appendix F, “Machine limits and rules,” on page 285.
v The current number of unit addresses defined on channel path types CNC, CTC, FC, and FCV are
displayed. For more information about the maximum unit addresses supported for a channel path
type, see Appendix F, “Machine limits and rules,” on page 285. IOCP might generate more unit
addresses for a spanned FC channel path (PATH keyword in the CHPID statement) than for an
unspanned channel path. See “FC control unit definition considerations:” on page 86. Also, IOCP might
generate more unit addresses for a shared CTC channel path than for an unshared channel path. See
“Shared ESCON CTC considerations” on page 25.
v The AID and PORT keyword values defined on channel path type CIB or CS5.
v The PCHID and PORT keyword values defined on channel path type CL5.
v The first FID number defined on a PCHID, PCHID-PORT, or VCHID.
The following channel configuration discrepancies can exist for a channel path:
YES - It exceeds max number of FICON+OSA | ISC-3 | CFS+CFR | CBS+CBR | CFP | CBP | ICP |
FCV | OSA channel paths allowed.
This discrepancy indicates that the channel path type has exceeded the maximum number of
channel paths allowed for your CPC for the indicated channel hardware type. This discrepancy
exists for all higher numbered channel paths of the same channel hardware type.
YES - It exceeds max number of CIB Adapter IDs allowed
This discrepancy indicates that the CIB Adapter ID (AID) has exceeded the maximum number
allowed for your CPC. This discrepancy exists for all other channel paths defined on the same
AID and for all channel paths on higher numbered AIDs.
In addition to the Channel Configuration Discrepancy report column, IOCP prints an informational line
after a channel path that has certain discrepancies. These informational lines are as follows:

PCHID xxx | CHPID xx in CSS y has exceeded the maximum number of FICON+OSA |
CF Hardware | ISC-3 | CFS+CFR | CBS+CBR | CFP | CBP | ICP | FCV | OSA channel
paths allowed by your processor.
AID xx has exceeded the maximum number of Adapter IDs allowed by your processor
The following channel path types are associated with the channel hardware types listed in the first
informational line:
v FICON+OSA = FC, FCV, FCP, OSC, OSD, OSE, OSM, OSN, and OSX
v CF Hardware = CFS, CFR, CFP, CBS, CBR, and CBP
v ISC-3 = CFS, CFR, and CFP
v CFS+CFR = CFS and CFR
v CBS+CBR = CBS and CBR
v OSA = OSC, OSD, OSE, OSM, OSN, and OSX
IOCP lists caution message ICP055I at the end of this report when any channel configuration
discrepancies exist.
The totals for the internal channel path types and internal FID types your CPC supports are also
included.

Sample PCHID summary report
TIME 15.11 DATE 17.094 REPORT BY ICP IOCP VERSION 5 RELEASE 4.0 PCHID SUMMARY REPORT PAGE 007

ID1=SYSTEM1
ID2=HCDI.IODF02.PROD - 2017-04-03 13:54
AID-PORT CSS NUMBERS

PCHID CHPID TYPE SWITCH |0|1|2|3|4|5| PNETID CHANNEL CONFIGURATION DISCREPANCY OR COMMENT
_____ _____ ____ ______ _____________ ________________ ____________________________________________________________
100 10 FCP N/A |0| | | | | | ---- NO - SUBCHANNELS = 0

101(M) 11 FC NONE |0| | | | | | ---- NO - UNIT ADDRESSES = 192
104 10 FC NONE | |1| | | | | ---- NO - UNIT ADDRESSES = 8
105 12 FC NONE |0| | | | | | ---- NO - UNIT ADDRESSES = 16
108 13 FC 25 |0| | | | | | ---- NO - UNIT ADDRESSES = 192
109 41 FC NONE | |1| | | | | ---- NO - UNIT ADDRESSES = 192
140 40 FC 25 | |1| | | | | ---- NO - UNIT ADDRESSES = 192

160 30 OSD N/A |0| | | | | | ---- NO - SUBCHANNELS = 240

OSDNET
161 31 OSE N/A | | | |3| | | ---- NO
200 N/A ROCE N/A | | | | | | | NET1 NO - FID = 0001

NETWORK2
204 N/A ZEDC N/A | | | | | | | ---- NO - FID = 0180
208 N/A RCE N/A | | | | | | | ---- NO - FID = 2100
210-1 N/A ROC2 N/A | | | | | | | ---- NO - FID = 1300

210-2 N/A ROC2 N/A | | | | | | | NET1 NO - FID = 1301
218-2 N/A HYL N/A | | | | | | | ---- NO - FID = 3500
280 80 FC 45 | | | |3| | | ---- NO - UNIT ADDRESSES = 192

281 90 FC 36 | | | |3| | | ---- NO - UNIT ADDRESSES = 192
290-1 F8 CL5 N/A |0| | | | | | ---- NO - DEFINED ON PCHID 290, PORT 1

290-2 F9 CL5 N/A |0| | | | | | ---- NO - DEFINED ON PCHID 290, PORT 2
N/A FC ICP N/A |0|1| | | | | ---- NO

N/A FD ICP N/A |0|1| | | | | ---- NO
7E5(V) N/A ISM N/A | | | | | | | ABCDNETWORK NO - FID = 4000
7F0(V) FE IQD N/A |0|1| | | | | ---- NO - SUBCHANNELS = 192, TOTAL = 592
7F1(V) FF IQD N/A |0|1| |3| | | ---- NO - SUBCHANNELS = 400, TOTAL = 592
08-2 F2 CIB N/A |0| | | | | | ---- NO - DEFINED ON AID 08, PORT 2
12-1 F3 CS5 N/A |0| | | | | | ---- NO - DEFINED ON AID 12, PORT 1
TOTALS FOR CARD TYPES:

______________________
FICON HCA CLR ICA INTERNAL

FC FCP CIB CL5 CS5 ICP IQD
__ ___ ___ ___ ___ ___ ___
12 1 1 2 1 2 2
OSA
OSC OSD OSE OSM OSX
___ ___ ___ ___ ___
0 1 1 0 0
FID
HYL ISM RCE ROCE ROC2 ZEDC
___ ___ ___ ____ ____ ____
1 1 1 1 1 1
Figure 23. Sample PCHID summary report
Channel path identifier (CHPID) summary report

The Channel Path Identifier (CHPID) Summary Reports list the channel path identifiers defined in each
CSS in the IOCDS. The report name in the heading states which CSS is being reported and to which CSS
the CHPID numbers belong. Every defined CSS is reported separately. For each CHPID, the report shows
the following:
with the channel path. VCHID numbers are indicated by '(V)'.
v Channel path type
v Channel path parameter (CHPARM)
v Logical partition (LP) access method for the channel path

v MIF image ID numbers of logical partitions that the channel path can access.
Figure 24 on page 161 shows an example of the CHPID Summary Report.
SWITCH keyword is allowed but not specified for a channel path, then the report column contains NONE.
If the CHPARM keyword is not allowed for the channel path type, the report column contains N/A. If
CHPARM is allowed but not specified or has a value of 00, either NONE or 00 is shown, depending on the
channel path type.
The LP ACCESS report column shows the logical partition access method for the channel path.
v DED is reported if the channel path is dedicated to a single logical partition
v REC is reported if the channel path is defined as reconfigurable
v SHARED is reported if the channel path is defined as shared
v SPANNED is reported if the channel path is defined as spanned.
The MIF IMAGE ID NUMBERS report column shows the MIF image IDs of logical partitions that the
CHPID can access in the CSS.
v For each partition in the access list of the CHPID, the MIF image ID number column contains an A for
the corresponding partition.
v For each partition in the candidate list but not the access list of the CHPID, the MIF image ID number
v For all partitions not in the access or candidate lists of the CHPID, the corresponding MIF image ID
number columns are left blank.
Several legends are located at the end of the CHPID Summary Report. The first set of legends describes
the entries in the CHPARM report column for the channel path types that support the CHPARM
keyword. The last legend describes the entries in the MIF IMAGE ID NUMBERS report column.
At the end of the CHPID Summary Report, the totals are listed for each channel path type your CPC
supports in the CSS.

Sample CHPID summary report
TIME 19.29 DATE 10.173 REPORT BY ICP IOCP VERSION 3 RELEASE 1.0 CSS 0: CHPID SUMMARY REPORT PAGE 008

ID1=SYSTEM1
ID2=HCDI.IODF02.PROD - 2005-11-10 13:54
AID-PORT LP MIF IMAGE ID NUMBERS

CHPID PCHID TYPE SWITCH CHPARM ACCESS |1|2|3|4|5|6|7|8|9|A|B|C|D|E|F|
_____ _____ ____ ______ ______ _______ _______________________________
10 100 CVC NONE N/A REC |C|C|C|C|A|C|C|C|C|C|C|C|C|C|C|

11 101 CNC NONE NONE SHARED |C|C|C|C|C|A|A|C|C|C|C|C|C|C|C|
12 103 CVC NONE N/A DED | | | | | |A| | | | | | | | | |
13 104 CNC 25 NONE SHARED |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
18 106 CNC NONE NONE SHARED |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
19 107 CNC NONE NONE SHARED |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
30 160 OSD N/A 00 SHARED |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
F2 08-2 CIB N/A N/A SHARED |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
FC N/A ICP N/A N/A SPANNED |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
FD N/A ICP N/A N/A SPANNED |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
FE N/A IQD N/A 00 SPANNED |C|A| | | | |A| | |A|C| | | | |
FF N/A IQD N/A 00 SPANNED |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
LEGEND FOR CHPARM FIELD FOR CHANNEL TYPES CNC, FCV, AND FC: LEGEND FOR CHPARM FIELD FOR CHANNEL TYPE IQD:
___________________________________________________________ _____________________________________________
01 - CHANNEL IS MANAGED BY DYNAMIC CHPID MANAGEMENT 0x - 8K MTU AND 16K FRAME SIZE 4x - 16K MTU AND 24K FRAME SIZE
8x - 32K MTU AND 40K FRAME SIZE Cx - 56K MTU AND 64K FRAME SIZE
x0 - REGULAR IQD CHPID x2 - IQD CHPID FOR IEDN
LEGEND FOR CHPARM FIELD FOR CHANNEL TYPES OSD, OSM, OSX: x4 - IQD CHPID BRIDGED TO EXTERNAL NETWORK
________________________________________________________
00 - MAXIMUM OF 160 IP STACKS AND 480 SUBCHANNELS

02 - GREATER THAN 160 IP STACKS AND 480 SUBCHANNELS LEGEND FOR CHPARM FIELD FOR CHANNEL TYPE OSC:
_____________________________________________
00 - 3270 CONSOLE FUNCTION

40 - 3215 TPF CONSOLE FUNCTION
______________________________________

TOTALS FOR CHPIDS FOR CSS 0:

____________________________
PARALLEL ESCON FICON COUPLING

CBY CVC CNC CTC FC CFP CIB ICP
___ ___ ___ ___ __ ___ ___ ___
0 2 4 0 0 0 1 2
NETWORK / QDIO
OSC OSD OSE OSM OSN OSX IQD FCP
___ ___ ___ ___ ___ ___ ___ ___
0 1 0 0 0 0 2 0
Figure 24. Sample CHPID summary report (Part 1 of 2)

TIME 19.29 DATE 10.173 REPORT BY ICP IOCP VERSION 3 RELEASE 1.0 CSS 1: CHPID SUMMARY REPORT PAGE 009

ID1=SYSTEM1
ID2=HCDI.IODF02.PROD - 2005-11-10 13:54
AID-PORT LP MIF IMAGE ID NUMBERS

CHPID PCHID TYPE SWITCH CHPARM ACCESS |1|2|3|4|5|6|7|8|9|A|B|C|D|E|F|
_____ _____ ____ ______ ______ _______ _______________________________
10 102 CTC NONE N/A SHARED | | | | |A| | |A| | | | | | | |

20 180 CNC 25 01 SHARED |C|C|C|C|C|C|C|C|C|C|C|C|C|C|C|
21 141 FC 35 NONE SHARED |A| | | |A| | |A| | | | | | | |
40 140 FC 26 NONE SHARED |A|A| |A|A| | |A| | |A| | | |A|
41 105 CNC NONE NONE SHARED |A|A| |A|A| | |A| | |A| | | |A|
FC N/A ICP N/A N/A SPANNED |A|A| |A|A| | |A| | |A| | | |A|
FD N/A ICP N/A N/A SPANNED |A|A| |A|A| | |A| | |A| | | |A|
FE N/A IQD N/A 00 SPANNED |C|A| |C|C| | |A| | |C| | | |C|
FF N/A IQD N/A 00 SPANNED |A|A| |A|A| | |A| | |A| | | |A|
LEGEND FOR CHPARM FIELD FOR CHANNEL TYPES CNC, FCV, AND FC: LEGEND FOR CHPARM FIELD FOR CHANNEL TYPE IQD:
___________________________________________________________ _____________________________________________
01 - CHANNEL IS MANAGED BY DYNAMIC CHPID MANAGEMENT 00 - 8K MTU AND 16K FRAME SIZE
40 - 16K MTU AND 24K FRAME SIZE
80 - 32K MTU AND 40K FRAME SIZE
LEGEND FOR CHPARM FIELD FOR CHANNEL TYPES OSD, OSM, OSX: C0 - 56K MTU AND 64K FRAME SIZE
________________________________________________________
00 - MAXIMUM OF 160 IP STACKS AND 480 SUBCHANNELS

02 - GREATER THAN 160 IP STACKS AND 480 SUBCHANNELS LEGEND FOR CHPARM FIELD FOR CHANNEL TYPE OSC:
_____________________________________________
00 - 3270 CONSOLE FUNCTION

40 - 3215 TPF CONSOLE FUNCTION
______________________________________

TOTALS FOR CHPIDS FOR CSS 1:

____________________________
PARALLEL ESCON FICON COUPLING

CBY CVC CNC CTC FC CFP CIB ICP
___ ___ ___ ___ __ ___ ___ ___
0 0 2 1 2 0 0 2
NETWORK / QDIO
OSC OSD OSE OSM OSN OSX IQD FCP
___ ___ ___ ___ ___ ___ ___ ___
0 0 0 0 0 0 2 0
Figure 25. Sample CHPID summary report (Part 2 of 2)
I/O device report

The I/O Device Report lists the I/O devices defined in the IOCDS. Devices are reported by the type of
control unit to which they attach. Within a group of devices that attach to the same type of control unit,
devices are sorted by device number. All the devices defined by the same IODEVICE statement are
reported together. For each IODEVICE statement, the report shows the following:
v Device numbers of the devices
v Subchannel set IDs in which the devices are defined
v Unit address of the devices
v Device type and model
v CSS IDs to which the devices have access
v Physical control unit numbers associated with the CSS ID and to which the devices are attached
v Paths that attach to the physical control unit
v MIF image ID numbers of logical partitions which the channel path can access

v Any comments for the devices.
Figure 26 on page 165 shows an example of the I/O Device Report.
If a device is in a subchannel set other than 0, the DEVICE NUMBERS report column for the appropriate
CSSs contains (n). The n represents the appropriate subchannel set number. If a device is in subchannel
set 1 for a CSS, the DEVICE NUMBERS report column for the appropriate CSSs contains (1). For devices
in subchannel set 0 for a CSS, no special indication is present.
The path listed is the composite path that the CPC uses when communicating with the device. This
composite path (chpid.link.cuadd) includes the channel path, the link address, and the control unit logical
address. If you do not specify LINK but you do specify CUADD, two asterisks (**) are displayed for
LINK. If a control unit is managed (PATH keyword on CNTLUNIT statement), two asterisks (**) are
displayed for the path.
SWITCH keyword is allowed but not specified for a channel path, the report column contains NONE. If the
control unit path is a managed path (PATH keyword on CNTLUNIT statement), the report column
contains YES
The MIF IMAGE ID NUMBERS report column shows the MIF image ID numbers of logical partitions
which the channel path can access. The following information is reported:
v For each partition in the access list of the channel path, the MIF image ID column contains an A for the
corresponding partition.
v For each partition in the candidate list but not the access list of the channel path, the MIF image ID
v For all partitions not in the access or candidate lists of channel path, the corresponding MIF image ID
columns are left blank.
You can use the PART, PARTITION, or NOTPART keyword in the IODEVICE statement to define a
device candidate list that includes or excludes LP access at the device level. The following information is
reported in the MIF IMAGE ID NUMBERS report column for a device candidate list:
v For the first report line for a device, the MIF image ID column contains an asterisk (*) for each
partition in the device candidate list.
v A minus sign (-) for an LP indicates that the device candidate list excludes LP access to the device even
though the LP has access to a channel path assigned to the device.
v A plus sign (+) for an LP indicates that the device candidate list includes LP access to the device even
though the LP currently does not have access to a channel path assigned to the device. The LP can
access only the device following a dynamic I/O configuration change that adds the LP to the candidate
list of the channel path.
Notes:
1. For each device that has a plus sign (+) in the MIF IMAGE ID NUMBERS column, ensure that you
specified the correct LPs in the PARTITION or NOTPART keyword on the IODEVICE statement. The
device does not have access to those LPs, even though the channel path does have access, until a
dynamic I/O configuration change is made.
2. If the device has a device candidate list that is equivalent to the one it would have for a default, a
plus sign (+) is not shown for LPs included in the device candidate list. A plus sign (+) is shown only
if the device candidate list was specified, is not equivalent to the default one, and included an LP for
which the path does not have access.
The COMMENT report column contains information that applies to the devices. The following
information is reported:
v STADET if the device has value STADET=Y

v TIMEOUT if the TIMEOUT keyword is not ignored for the device and has value TIMEOUT=Y
v Protocol value if the PROTOCL keyword is not ignored for the control units attached to the device as
follows:
– DCI for PROTOCL=D
– DS-3.0 for PROTOCL=S
– DS-4.5 for PROTOCL=S4
v TYPE value of 1 or 2 if the SHARED keyword is not ignored for the control units attached to the
device
v PREFERRED PATH if the channel path is the preferred channel path for the device
v MANAGED PATHS (x) if the control unit attached to the device has managed paths and where x is the
number of managed paths in the control unit.
A legend at the end of the I/O Device Report describes the entries in the MIF IMAGE ID NUMBERS
report column.

Sample I/O device report
TIME 19.29 DATE 10.173 REPORT BY ICP IOCP VERSION 3 RELEASE 1.0 I/O DEVICE REPORT PAGE 012

ID1=SYSTEM1
ID2=HCDI.IODF02.PROD - 2005-11-10 13:54
CONTROL PATH
DEVICE UNIT DEVICE UNIT CHPID.LINK.CUADD MIF IMAGE ID NUMBERS
NUMBERS ADDRS TYPE-MODEL CSS NUMBER |__.__.__| SWITCH |1|2|3|4|5|6|7|8|9|A|B|C|D|E|F| COMMENT
_________ _____ __________ ___ _______ ______ _______________________________ ____________________________
******************** The following devices are assigned to control unit type CFP ********************
------------------------------------------------------------------------------------------------------------------------------------
FFE0-FFE6 N/A CFP 0 | | | | | | | | | | | | | | | |
FFE0 F2 N/A |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
------------------------------------------------------------------------------------------------------------------------------------
FFF2-FFFF N/A CFP 0 | | | | | | | | | | | | | | | |
FFFE FC N/A |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
FD N/A |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
----------------------------------------------------------------------------------------------------
1 | | | | | | | | | | | | | | | |
FFFE FC N/A |A|A| |A|A| | |A| | |A| | | |A|
FD N/A |A|A| |A|A| | |A| | |A| | | |A|
------------------------------------------------------------------------------------------------------------------------------------
******************** The following devices are assigned to control unit type FCTC ********************
------------------------------------------------------------------------------------------------------------------------------------
3000 00 FCTC 1 |*|*|*|*|*|*|*|*|*|*|*|*|*|*|*| STADET
800 21.635D.14 35 |A| | | |A| | |A| | | | | | | |
------------------------------------------------------------------------------------------------------------------------------------
******************** The following devices are assigned to control unit type IQD ********************
------------------------------------------------------------------------------------------------------------------------------------
4200-420F 00-0F IQD 0 |*|*| | | | |*| | |*|*| | | | |
F000 FE N/A |C|A| | | | |A| | |A|C| | | | |
----------------------------------------------------------------------------------------------------
1 |*|*| |*|*| | |*| | |*| | | |*|
F000 FE N/A |C|A| |C|C| | |A| | |C| | | |C|
------------------------------------------------------------------------------------------------------------------------------------
4300-430F 00-0F IQD 0 |*|*|*|*|*|*|*|*|*|*|*|*|*|*|*|
F001 FF N/A |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
----------------------------------------------------------------------------------------------------
1 |*|*| |*|*| | |*| | |*| | | |*|
F001 FF N/A |A|A| |A|A| | |A| | |A| | | |A|
----------------------------------------------------------------------------------------------------
3 |*|*|*| | | | | | | | | | | | |
F001 FF N/A |A|A|A| | | | | | | | | | | | |
------------------------------------------------------------------------------------------------------------------------------------
******************** The following devices are assigned to control unit type OSA ********************
Figure 26. Sample I/O device report (Part 1 of 3)

TIME 19.29 DATE 10.173 PAGE 013
CONTROL PATH
_________ _____ __________ ___ _______ ______ _______________________________ ____________________________
------------------------------------------------------------------------------------------------------------------------------------
2000-200F 00-0F OSA 0 |*|*|*|*|*|*|*|*|*|*|*|*|*|*|*|
003 30 N/A |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
------------------------------------------------------------------------------------------------------------------------------------
20FE FE OSAD 0 |*|*|*|*|*|*|*|*|*|*|*|*|*|*|*|
003 30 N/A |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
------------------------------------------------------------------------------------------------------------------------------------
******************** The following devices are assigned to control unit type OSN ********************
------------------------------------------------------------------------------------------------------------------------------------
2801-2810 01-10 3745 3 |*|*|*| | | | | | | | | | | | |
004 31 N/A |A|A|A| | | | | | | | | | | | |
------------------------------------------------------------------------------------------------------------------------------------
2880-28BF 80-BF OSN 3 |*|*|*| | | | | | | | | | | | |
004 31 N/A |A|A|A| | | | | | | | | | | | |
------------------------------------------------------------------------------------------------------------------------------------
******************** The following devices are assigned to control unit type SCTC ********************
------------------------------------------------------------------------------------------------------------------------------------
100 00 SCTC 1 | | | | |*| | |*| | | | | | |*| STADET
002 10 NONE | | | | |A| | |A| | | | | | |+|
------------------------------------------------------------------------------------------------------------------------------------
101 01 SCTC 1 |*|*| |*| | | |*| | |*| | | |*| STADET
002 10 NONE |+|+| |+|-| | |A| | |+| | | |+|
------------------------------------------------------------------------------------------------------------------------------------
******************** The following devices are assigned to control unit type 2105 ********************
------------------------------------------------------------------------------------------------------------------------------------
600-67F 00-7F 3390B 0 |*|*|*|*|*|*|*|*|*|*|*|*|*|*|*| STADET
044 11.**.2 NONE |C|C|C|C|C|A|A|C|C|C|C|C|C|C|C|
13.D0.2 25 |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
** YES |C|C|C|C|C|C|C|C|C|C|C|C|C|C|C| MANAGED PATHS (2)
----------------------------------------------------------------------------------------------------
1 |*|*|*|*|*|*|*|*|*|*|*|*|*|*|*|
044 40.C0.2 26 |A|A| |A|A| | |A| | |A| | | |A|
41.**.2 NONE |A|A| |A|A| | |A| | |A| | | |A|
----------------------------------------------------------------------------------------------------
3 |*|*|*|*|*|*|*|*|*|*|*|*|*|*|*|
044 80.E0.2 45 |A|A|A| | | | | | | | | | | | |
90.E0.2 36 |A|A|A| | | | | | | | | | | | |
------------------------------------------------------------------------------------------------------------------------------------
680-6BF(1) 80-BF 3390A 0 |*|*|*|*|*|*|*|*|*|*|*|*|*|*|*| STADET
044 11.**.2 NONE |C|C|C|C|C|A|A|C|C|C|C|C|C|C|C|
13.D0.2 25 |A|A|A|A|A|A|A|A|A|A|A|A|A|A|A|
** YES |C|C|C|C|C|C|C|C|C|C|C|C|C|C|C| MANAGED PATHS (2)

TIME 19.29 DATE 10.173 PAGE 014
CONTROL PATH
_________ _____ __________ ___ _______ ______ _______________________________ ____________________________
1 |*|*|*|*|*|*|*|*|*|*|*|*|*|*|*|
044 40.C0.2 26 |A|A| |A|A| | |A| | |A| | | |A|
41.**.2 NONE |A|A| |A|A| | |A| | |A| | | |A|
----------------------------------------------------------------------------------------------------
(1) 3 |*|*|*|*|*|*|*|*|*|*|*|*|*|*|*|
044 80.E0.2 45 |A|A|A| | | | | | | | | | | | |
90.E0.2 36 |A|A|A| | | | | | | | | | | | |
------------------------------------------------------------------------------------------------------------------------------------
******************** The following devices are assigned to control unit type 3490 ********************
------------------------------------------------------------------------------------------------------------------------------------
1020-1027 20-27 3480 0 | | | | | | | | | | | | | | | | STADET,TIMEOUT,DS-4.5,TYPE 2
1111 10 NONE |C|C|C|C|A|C|C|C|C|C|C|C|C|C|C| PREFERRED PATH
12 NONE | | | | | |A| | | | | | | | | |
------------------------------------------------------------------------------------------------------------------------------------
1028-102F 28-2F 3480 0 | | | | | | | | | | | | | | | | STADET,TIMEOUT,DS-4.5,TYPE 2
1111 10 NONE |C|C|C|C|A|C|C|C|C|C|C|C|C|C|C|
12 NONE | | | | | |A| | | | | | | | | | PREFERRED PATH
------------------------------------------------------------------------------------------------------------------------------------

______________________________________
* - PARTITION IN DEVICE’S CANDIDATE LIST FOR THE CSS

A - PARTITION IN CHPID’S ACCESS AND CANDIDATE LISTS FOR THE CSS
C - PARTITION IN CHPID’S CANDIDATE LIST ONLY FOR THE CSS
BLANK - PARTITION NOT IN CHPID’S ACCESS OR CANDIDATE LIST FOR THE CSS OR NOT IN DEVICE’S CANDIDATE LIST FOR THE CSS
- - PARTITION IN CHPID’S ACCESS OR CANDIDATE LIST BUT NOTPART OR PARTITION KEYWORD ON IODEVICE EXCLUDED THIS PARTITION
FROM DEVICE’S CANDIDATE LIST FOR THE CSS
+ - PARTITION NOT IN CHPID’S ACCESS OR CANDIDATE LIST BUT NOTPART OR PARTITION KEYWORD ON IODEVICE INCLUDED THIS PARTITION
IN DEVICE’S CANDIDATE LIST FOR THE CSS
Channel path report

The Channel Path Report, like the PCHID Summary Report and CHPID Summary Report, lists the
channel paths defined in the IOCDS. The Channel Path Report shows the following information for each
channel path defined in the IOCDS:
with the channel path. VCHID numbers are indicated by '(V)'. If the MIXTYPE keyword was specified
for the PCHID, this is indicated by '(M)'.
v Channel path type
v Logical partition (LP) access method for the channel path
v Physical control unit numbers that attach to the channel path
v Device numbers that attach to the physical control units that attach to the channel path.
For each physical control unit that attaches to each channel path, the Channel Path Report shows the
following:
v Control unit number and type
v Control unit logical address (CUADD)
v Unit address ranges
v CSS IDs to which the channel path has access
v Link address.
All the devices defined by the same IODEVICE statement are reported together. For each IODEVICE
statement that is assigned to each physical control unit, the Channel Path Report shows the following:

v Device numbers from the IODEVICE statement assigned to this channel path
v Subchannel set IDs in which the devices are defined
v Unit addresses for the devices
v Device type
v Model.
See Figure 29 on page 169 for an example of the Channel Path Report.
SWITCH keyword is allowed but not specified for a channel path, the report column contains NONE.
The LP ACCESS report column shows the logical partition access method for the channel path.
v DED is reported if the channel path is dedicated to a single logical partition
v REC is reported if the channel path is defined as reconfigurable
v SHARED is reported if the channel path is defined as shared
v SPANNED is reported if the channel path is defined as spanned.
If a channel path is assigned to an I/O cluster (IOCLUSTER keyword), the I/O cluster name is reported
in the control unit columns.
For ICP channel paths, a CSS ID associated with the channel path and the CHPID number and CSS ID
that are associated with the connected channel path are reported in the control unit columns.
For CIB, CL5, and CS5 channel paths, a CSS ID associated with the channel path and the CHPID number
and CSS ID and CSYSTEM name associated with the connecting channel path are reported in the control
unit columns.
If the channel path type (TYPE keyword) does not permit logical addresses (CUADD keyword) to be
specified for physical control units, the report column contains N/A. If the channel path type permits
logical addresses but you did not specify any, the report column contains NONE.
If the channel path type (TYPE keyword) does not permit link addresses (LINK keyword) to be specified
for physical control units, the report column contains N/A. If the channel path type permits link addresses
but you did not specify any, the report column contains two asterisks (**).
If a device is in a subchannel set other than 0, (n) is contained in the DEVICE NUMBERS report column
for the appropriate CSSs. The n represents the appropriate subchannel set number. For devices in
subchannel set 0 for a CSS, no special indication is included.

Sample channel path report
TIME 16.39 DATE 17.164 REPORT BY ICP IOCP VERSION 5 RELEASE 4.0 CHANNEL PATH REPORT PAGE 016

ID1=SYSTEM1
ID2=HCDI.IODF02.PROD - 2017-04-03 13:54
AID-PORT LP CONTROL UNIT UA DEVICE UNIT DEVICE

PCHID CHPID TYPE SWITCH ACCESS NUMBER TYPE CUADD RANGE CSS LINK NUMBERS ADDRS TYPE MODEL
_____ _____ ____ ______ _______ _____________________ _____ ___ ____ _________ _____ ______________
100 10 FCP N/A REC

------------------------------------------------------------------------------------------------------------------------------------
101(M) 11 FC NONE SHARED
044 2105 2 00-BF 0 **
600-67F 00-7F 3390B
680-6BF(1) 80-BF 3390A
------------------------------------------------------------------------------------------------------------------------------------
104 10 FC NONE SHARED
002 SCTC NONE 00-07 1 **
100 00 SCTC
101 01 SCTC
------------------------------------------------------------------------------------------------------------------------------------
105 12 FC NONE DED
1111 3490 NONE 20-2F 0 **
1020-1027 20-27 3480
1028-102F 28-2F 3480
------------------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------
108 13 FC 25 SHARED
044 2105 2 00-BF 0 D0
600-67F 00-7F 3390B
680-6BF(1) 80-BF 3390A
------------------------------------------------------------------------------------------------------------------------------------
044 2105 2 00-BF 1 **
600-67F 00-7F 3390B
680-6BF 80-BF 3390A
------------------------------------------------------------------------------------------------------------------------------------
140 40 FC 25 SHARED
044 2105 2 00-BF 1 C0
600-67F 00-7F 3390B
680-6BF 80-BF 3390A
------------------------------------------------------------------------------------------------------------------------------------
141 21 FC 35 SHARED
800 FCTC 14 00-07 1 635D
3000 00 FCTC
-----------------------------------------------------------------------------------------------
801 2105 NONE 00-FF 1 635E
-----------------------------------------------------------------------------------------------
802 2105 NONE 00-FF 1 635F
-----------------------------------------------------------------------------------------------
803 2105 NONE 00-FF 1 6360
-----------------------------------------------------------------------------------------------
804 2105 NONE 00-FF 1 6361
------------------------------------------------------------------------------------------------------------------------------------
160 30 OSD N/A SHARED
003 OSA NONE 00-FE 0 N/A
2000-200F 00-0F OSA
20FE FE OSAD
------------------------------------------------------------------------------------------------------------------------------------
161 31 OSE N/A SHARED
004 OSN N/A 00-FE 3 N/A
2801-2810 01-10 OSA
2880-28BF 80-BF OSA
------------------------------------------------------------------------------------------------------------------------------------
180 20 FC 25 SHARED *** IOCLUSTER = SYSPLEX1 ***
------------------------------------------------------------------------------------------------------------------------------------
280 80 FC 45 SHARED
044 2105 2 00-BF 3 E0
600-67F 00-7F 3390B
680-6BF(1) 80-BF 3390A
------------------------------------------------------------------------------------------------------------------------------------
281 90 FC 36 SHARED
044 2105 2 00-BF 3 E0
600-67F 00-7F 3390B
680-6BF(1) 80-BF 3390A
------------------------------------------------------------------------------------------------------------------------------------
290-1 F8 CL5 N/A SHARED *** IS IN CSS 0 AND CONNECTS TO CL5 CHPID 80 IN CSS 5 ON SYSTEM CPC3 ***
------------------------------------------------------------------------------------------------------------------------------------
290-2 F9 CL5 N/A SHARED *** IS IN CSS 0 AND CONNECTS TO CL5 CHPID 81 IN CSS 5 ON SYSTEM CPC3 ***
------------------------------------------------------------------------------------------------------------------------------------
Figure 29. Sample channel path report (Part 1 of 2)

TIME 16.39 DATE 17.164 PAGE 017
AID-PORT LP CONTROL UNIT UA DEVICE UNIT DEVICE

PCHID CHPID TYPE SWITCH ACCESS NUMBER TYPE CUADD RANGE CSS LINK NUMBERS ADDRS TYPE MODEL
_____ _____ ____ ______ _______ _____________________ _____ ___ ____ _________ _____ ______________
N/A FC ICP N/A SPANNED *** IS IN CSS 0 AND CONNECTS TO ICP CHPID FD IN CSS 1 ***
FFFE CFP N/A N/A 0 N/A
FFF2-FFFF N/A CFP
-----------------------------------------------------------------
1 N/A
FFF2-FFFF N/A CFP
------------------------------------------------------------------------------------------------------------------------------------
N/A FD ICP N/A SPANNED *** IS IN CSS 0 AND CONNECTS TO ICP CHPID FC IN CSS 0 ***
FFFE CFP N/A N/A 0 N/A
FFF2-FFFF N/A CFP
-----------------------------------------------------------------
1 N/A
FFF2-FFFF N/A CFP
------------------------------------------------------------------------------------------------------------------------------------
7F0(V) FE IQD N/A SPANNED
F000 IQD NONE 00-FF 0 N/A
4200-420F 00-0F IQD
-----------------------------------------------------------------
1 N/A
4200-420F 00-0F IQD
------------------------------------------------------------------------------------------------------------------------------------
7F1(V) FF IQD N/A SPANNED
F001 IQD NONE 00-FF 0 N/A
4300-430F 00-0F IQD
-----------------------------------------------------------------
1 N/A
4300-430F 00-0F IQD
-----------------------------------------------------------------
3 N/A
4300-430F 00-0F IQD
------------------------------------------------------------------------------------------------------------------------------------
08-2 F2 CIB N/A SHARED *** IS IN CSS 0 AND CONNECTS TO CIB CHPID F0 IN CSS 1 ON SYSTEM CPC2 ***
FFE0 CFP N/A N/A 0 N/A
FFE0-FFE6 N/A CFP
------------------------------------------------------------------------------------------------------------------------------------
12-1 F3 CS5 N/A SHARED *** IS IN CSS 0 AND CONNECTS TO CS5 CHPID F0 IN CSS 2 ON SYSTEM CPC3 ***
------------------------------------------------------------------------------------------------------------------------------------
Figure 30. Sample channel path report (Part 2 of 2)
Conditions for producing reports

The z/OS and VSE versions of IOCP produce reports, if no error occurs, as follows:
v IOCP generates an IOCDS. The reports show the configuration data in the IOCDS that is based on the
input statements.
v Code the REPORT option in the PARM parameter of the EXEC statement. The reports show the
configuration data contained in each IOCDS that you specified on REPORT=.
The VM version of IOCP produces reports, if no error occurs, as follows:

v IOCP generates an IOCDS in response to the CMS IOCP command with IOCDS generation options.
The reports show the configuration data in the IOCDS that IOCP generates from the input statements.
v Use configuration report options (RPTAx or RPTA) in the CMS IOCP command. The reports show the
configuration data contained in each specified IOCDS.
The stand-alone version of IOCP produces reports, if no error occurs, as follows:

v The Send output to a printer option was selected for the Build data set action. The reports show the
configuration data generated by the build IOCDS process.
v The Print data set report action was selected. The reports show the configuration data contained in the
selected IOCDS.

Chapter 8. IOCP messages
IOCP messages (ICP prefix)
This topic describes the messages produced by all versions (z/OS, z/VM, z/VSE, and stand-alone) of ICP
IOCP. The messages are described in alphanumeric sequence by message identifier.
Message ICP050D is sent to the z/OS or VSE system console and requests action by the system operator.
Messages ICP056I and ICP057I are sent to the z/OS or VSE system console and the output printer for the
IOCP programmer. All other messages are sent to the output printer or card-image file for the IOCP
programmer or to the system console for the IOCP operator.
The message format is:

ICPnnns text
nnn Is the message serial number.
s Is the type code:
A Action. The operator must perform a specific action.
D Decision. The operator must choose an alternative.
I Information. No operator action is required. Most information messages are for the IOCP
programmer.
W Wait. Processing stops until an action is determined and performed.
text Is the message text.
ICP000I UNDEFINED MESSAGE CODE. CODE=xxxxxxxx HEX

Explanation: IOCP has generated message code xxxxxxxx for which a message has not been defined. A logical error
has occurred in IOCP.
System action: IOCP action is unpredictable.
Programmer response: Rerun the IOCP job. If the problem occurs again, save the console sheet and all associated
output. Report the problem to your service representative.
ICP050D SHOULD jobname WRITE TO LEVEL Ax IOCDS? REPLY 'YES', 'NO' OR 'CANCEL'
Explanation: A job with name jobname is running IOCP with a WRTCDS option other than NO in the PARM
parameter of the EXEC statement. IOCP sends this message to request permission from the z/OS or VSE system
operator for the job to write the Ax IOCDS to the Support Element hard disk.
z/OS users can prevent this message from being sent. See “Authorizing an IOCDS write” on page 112.
Note: If two (or more) jobs are allowed to update concurrently the Ax IOCDS, the outcome might be an IOCDS that
is logically inconsistent with the input from any one job. Using this IOCDS at power-on reset might produce
undesirable results.
System action: If you enter YES, the job continues running IOCP and, if no errors are encountered, replaces the
input/output configuration data in the Ax IOCDS on the hard disk with the input/output configuration data it
generates. Use Input/Output Configuration window of the Support Element to determine if the Ax IOCDS is
write-protected.
If you enter NO, the job continues generating input/output configuration data in storage and to produce reports, but
the job does not replace the input/output configuration data in the Ax IOCDS on the SE hard disk.
If you enter CANCEL, the job ends immediately and receives system code X'222'.

ICP051I
Operator response: Contact your system programmer and respond as instructed.

Programmer response: Follow your site procedures and advise the operator.
Routing codes: 1 and 11
Descriptor code: 7

Explanation: IOCP version nn.nn.nnn has terminated for one of the following reasons:
Code Description
01 UNKNOWN PARM KEYWORD VALUE - yyyyyyyy. An unknown keyword value was found beginning with
the first character of string yyyyyyyy in the PARM parameter of the EXEC statement. (This code can result
from an extra left parenthesis after the REPORT keyword.)
02 DUPLICATE PARM KEYWORD - yyyyyyyy. A duplicate keyword was found beginning with the first
character of string yyyyyyyy in the PARM parameter of the EXEC statement.
03 UNKNOWN PARM KEYWORD - yyyyyyyy. An unknown keyword was found beginning with the first
character of string yyyyyyyy in the PARM parameter of the EXEC statement.
04 INVALID LINECOUNT VALUE. The keyword value on the LINECOUNT keyword in the PARM parameter
of the EXEC statement was not within the allowed range.
05 CONFLICTING PARM KEYWORDS. Conflicting keywords were found in the PARM parameter of the EXEC
statement.
06 PARM DELIMITER ERROR - yyyyyyyy. An extra or missing comma was found beginning with the first
character of string yyyyyyyy in the PARM parameter of the EXEC statement. This code can result from
unbalanced parentheses around the REPORT parameters or an incorrect report keyword value.
07 One or more errors were found in the input card-image statements.
08 An error was found while IOCP was generating the IOCDS in storage.
09 A REPORT or WRTCDS option has been specified but ICP IOCP only reads or writes an IOCDS in a CPC
that supports ICP IOCP. If your CPC supports ICP IOCP, the IOCP release level you used does not support
your CPC; you must use an IOCP level that supports your CPC.
10 IOCP was unable to connect to the VSE facility which reads and writes an IOCDS for one of the following
reasons:
v The facility is in use by another job. Wait a few minutes and rerun the job.
v The facility is unavailable due to a or VSE error.
v VSE is a guest under VM. VSE does not support reading or writing an IOCDS as a guest under VM. Use
the VM version of IOCP.
11 TOO MANY VM LEVELS. You attempted to read or write an IOCDS and multiple levels of VM are present.
For z/OS, reading and writing an IOCDS is supported when z/OS is a guest of VM and only one level of
VM is present. For VM, reading and writing an IOCDS is supported only when performed from a first-level
VM.
12 IOCP was not authorized to write an IOCDS. The operator replied No to message ICP050D.
13 IOCP was not authorized to write an IOCDS. You are not authorized to update the CBD.CPC.IOCDS
resource of the FACILITY class in the installed security product. (See “Authorizing an IOCDS write” on page
112 for more information.)
14 PARM KEYWORD MISSING. Not all required execution keywords were specified. You specified keyword
CHECKCPC=NO in the PARM parameter of the EXEC statement or CMS option NOCHKCPC to write an
IOCDS in preparation of an upgrade, but you did not specify a keyword or option to write the IOCDS. You
must specify a keyword value other than WRTCDS=NO or NOWRTCDS.
15 INSUFFICIENT STORAGE. IOCP could not obtain enough storage by using the GETMAIN statement (for
z/OS and z/VM) or the GETVIS statement (for VSE).

ICP052I • ICP055I
4x IOCP encountered an error while reading the Ax (where x is 0, 1, 2, or 3) IOCDS from, or writing the Ax
IOCDS to the hard disk. See the listing for related error messages.
System action: If IOCP encountered an error while doing anything other than reading an IOCDS, it ends with
return code 8. If IOCP encountered an error while reading an IOCDS, it ends the read operation for that IOCDS.
However, IOCP tries to read each additional IOCDS that was specified before it ends with return code 8.
Programmer response: Ensure that the PARM parameter of the EXEC statement is coded correctly. For errors, save
the output listings and report the problem to the service representative.
ICP052I DCB DID NOT OPEN-SYSIN DD STATEMENT MAY BE MISSING/INVALID

Explanation: The device control block for the data set specified in the SYSIN DD statement did not open when
IOCP sent the OPEN statement.
System action: IOCP ends.
Programmer response: Ensure that the DD statement for the SYSIN data set is valid.
ICP053I IOCP VERSION nn.nn.nnn FAILED. INSUFFICIENT STORAGE

Explanation: IOCP version nn.nn.nnn could not obtain enough storage with the GETMAIN statement (for z/OS and
z/VM) or the GETVIS statement (for VSE). If you are reporting on multiple IOCDSs, IOCP could not obtain enough
storage to report on any of the IOCDSs. IOCP performs multiple storage allocations. If the first storage request fails,
only this message is sent. If the second storage request fails, other messages are sent to the output data set.
System action: IOCP ends with a return code of 16.
Programmer response: For z/OS users, increase the region size. To determine the region size to use, see “IOCP
storage requirements” on page 112.
For z/VM users, define more virtual storage, perform an IPL of CMS, and issue the IOCP command again. For more
information, see “IOCP storage requirements” on page 112.
For z/VSE users, increase the partition size. Also, specify SIZE=AUTO on the EXEC statement of the job to allow
IOCP to use the remaining partition size for GETVIS. For more information, see “IOCP storage requirements” on
page 112.
Routing codes: 11
Descriptor code: 6
ICP054I **WARNING** IOCP CANNOT DETERMINE LIMITS FOR PROCESSOR. CODE=xxxxxxxx

Explanation: IOCP does not recognize the CPC that it is running on. A logical error has occurred in IOCP. IOCDSs
written on this CPC might not be valid for this CPC.
System action: IOCP sends this warning message after the last statement has been processed, defaults to a set of
predetermined machine limits and rules, and then continues processing.
Programmer response: To determine which machine limits and rules IOCP used as defaults, note the value of the
SYSTEM= keyword in the editor or from the report headings and then see Appendix F, “Machine limits and rules,”
on page 285. You can define all channel path types with IOCP. Ensure that the machine limits, channel limits, and
rules are valid for your CPC before using the IOCDS for a power-on reset. Report the problem to your service
representative.
ICP055I **CAUTION** CHANNEL CONFIGURATION CONTAINS DISCREPANCIES

Explanation: Your channel configuration contains one or more discrepancies. Discrepancies are channel
configuration conditions that cannot exist on your CPC. The discrepancies are reported for each channel path in the
PCHID Summary Report. See “PCHID summary report” on page 156 for a description of the channel configuration
discrepancies.
System action: IOCP prints this message at the end of the PCHID Summary Report and continues processing.
Programmer response: Change your channel configuration to correct the discrepancies. For an explanation of the
channel configuration rules for your CPC, see Appendix F, “Machine limits and rules,” on page 285.
Chapter 8. IOCP messages 173

ICP056I • ICP058I
ICP056I IOCP JOB jobname FAILED. DID NOT BUILD LEVEL Ax IOCDS.
Explanation: A job with name jobname was running IOCP with a WRTCDS option other than NO in the PARM
parameter of the EXEC statement. However, IOCP did not write the IOCDS. One of the following situations occurred:
v An error was detected in the IOCP input statements
v The z/OS or VSE system operator previously replied NO to message ICP050D
v The CBD.CPC.IOCDS resource of the FACILITY class has been defined to an installed security product and the
z/OS security administrator has not authorized you to update the CBD.CPC.IOCDS resource. See “Authorizing an
IOCDS write” on page 112.
IOCP sends this message to inform the programmer and the z/OS or VSE system operator (who previously
responded to message ICP050D) that this job failed to build the Ax IOCDS.
For z/OS users, IOCP only sends this message to the programmer (routing code 11) if the CBD.CPC.IOCDS resource
of the FACILITY class has been defined to an installed security product.
System action: IOCP ends the job with return code 8.
Operator response: Tell your system programmer that the job failed.
Programmer response: Review the messages in the job listing and take the necessary actions.
Descriptor code: 6
ICP057I IOCP JOB jobname SUCCESSFUL. LEVEL Ax IOCDS REPLACED.

Explanation: A job with name jobname was running IOCP with a WRTCDS option other than NO in the PARM
parameter of the EXEC statement. IOCP sends this message to inform the programmer and the z/OS or VSE system
operator (who previously granted permission to write the IOCDS in response to message ICP050D) that this job
successfully replaced the Ax IOCDS.
For z/OS users, IOCP only sends this message to the programmer (routing code 11) if the CBD.CPC.IOCDS resource
of the FACILITY class has been defined to an installed security product and the z/OS security administrator has
authorized you to update the CBD.CPC.IOCDS resource.
System action: IOCP ends the job with return code 0, 2, or 4.
Operator response: Follow your site procedures.
Programmer response: No response is required. This message is an informational message only.
Descriptor code: 6
ICP058I IOCP DID NOT WRITE IOCDS Ax

Explanation: IOCP was requested to write the Ax IOCDS to the hard disk. However, IOCP did not write the IOCDS.
One of the following situations occurred:
v An error was detected in the IOCP input statements
v The z/OS or VSE system operator previously replied NO to message ICP050D
v The CBD.CPC.IOCDS resource of the FACILITY class has been defined to an installed security product and the
z/OS security administrator has not authorized you to update the CBD.CPC.IOCDS resource. See “Authorizing an
IOCDS write” on page 112.
v IOCP received an unexpected response code or return code attempting to write the IOCDS.
This message is an informational message only.
System action: IOCP prints this informational message at the end of the configuration reports. IOCP continues
processing.
Programmer response: Do one of the following actions:
v Correct the IOCP input statements

ICP059I • ICP062I
v Have the z/OS or VSE system operator reply YES to message ICP050D
v Have the z/OS security administrator authorize you to update the CBD.CPC.IOCDS resource of the FACILITY
class.
v Perform the programmer response associated with the unexpected response code or return code.
ICP059I IOCP SUCCESSFULLY WROTE IOCDS Ax

Explanation: IOCP was requested to write the Ax IOCDS to the Support Element hard disk. IOCP successfully
wrote the IOCDS. This message is an informational message only.
System action: IOCP prints this informational message at the end of the configuration reports. IOCP continues
processing.
Programmer response: No response is required.
ICP060I IOCP COULD NOT OPEN OR WRITE TO OUTPUT DATASET

Explanation: IOCP could not open or write to the output data set specified in the SYSPRINT DD statement. The
SYSPRINT DD statement might be missing.
System action: IOCP ends with return code of 12.
Programmer response: Ensure that the DD statement for the SYSPRINT data set exists and is valid.
Routing codes: 11
Descriptor code: 6
ICP061I **WARNING** EXECUTION PARAMETER LPAR=YES IS INVALID

Explanation: You specified keyword LPAR=YES in the PARM parameter of the EXEC statement or CMS option
LPAR for an LPAR I/O configuration. However, IOCP supports only LPAR I/O configurations.
Note: Beginning with ICP IOCP 2.2.0, LPAR=YES and LPAR are disallowed.
System action: IOCP ignores the parameter and continues processing.
Programmer response: Do not specify LPAR=YES or LPAR.
ICP062I **WARNING** COMMENTS | MESSAGES REMOVED FROM CARD IMAGE FILE

Explanation: The card-image file was too large to read or write so stand-alone IOCP removed comments or IOCP
messages from the file to reduce its size. When reading a card-image file, stand-alone IOCP removes comments or
IOCP messages if necessary to reduce the file size so that it can read the entire file. When writing a card-image file to
the Support Element hard disk, stand-alone IOCP removes comments or IOCP messages if necessary to reduce the
file size so that it can be written.
The restricts the size of I/O configuration source (IOCS) files to 256,000 card records. Twenty additional records are
reserved for IOCP messages.
IOCP messages are messages that IOCP inserted into the card-image file during processing. If IOCP messages are
removed, the card-image file still contains the conditions that caused the messages but not the messages.
System action: IOCP inserts this message at the top of the card-image file and continues processing.
Programmer response: Reduce the size of the IOCS file you selected and then retry the IOCP function. To reduce
the size of an IOCS file perform one or more of the following actions:
v Remove comment cards from the card-image file
v Consolidate IOCP statements that are continued across multiple cards onto a minimum number of cards
v Combine individual IODEVICE statements that attach to the same control unit or units by using common UNIT
and MODEL keyword values. Do not combine IODEVICE statements if they have conflicting keyword values.

ICP063I • ICP065I
ICP063I ERRORS=YES | NO | NA, MESSAGES=YES | NO | NA, REPORTS PRINTED=YES | NO | NA,

IOCDS WRITTEN=YES | NO | NA, IOCS WRITTEN=YES | NO
Explanation: Stand-alone IOCP writes an I/O configuration source (IOCS) file for the Build data set and
Disassemble data set functions. IOCP always includes this message in the IOCS file. The status of each item is YES,
NO, or NA (not applicable for the requested IOCP function). YES indicates the following conditions:
ERRORS
IOCP processing resulted in error messages.
MESSAGES
IOCP processing resulted in caution or warning messages or both.
REPORTS PRINTED
IOCP successfully printed its reports.
IOCDS WRITTEN
IOCDS was successfully written to the Support Element hard disk.
IOCS WRITTEN
IOCS was successfully written to the Support Element hard disk.
The Support Element restricts the size of I/O configuration source (IOCS) files to 256,000 card records. Twenty
additional records are reserved for IOCP messages.
If an IOCS is too large to write, IOCP writes an IOCS containing only this message and the status for IOCS
WRITTEN is NO. Otherwise, this message is at the bottom of the IOCS file and the status for IOCS WRITTEN is YES.
System action: IOCP writes an IOCS containing this message and continues processing.
Programmer response: For ERRORS status of YES, correct all errors and retry the requested IOCP function.
For MESSAGES status of YES, ensure that the caution and warning messages do not indicate a condition that you
must correct even though an IOCDS might have been successfully written.
For IOCS WRITTEN status of NO, reduce the size of the IOCS file you selected and then retry the IOCP function. To
reduce the size of an IOCS file perform one or more of the following actions:
v Remove comment cards from the IOCS.
ICP064I **CAUTION** IOCDS IS UNUSABLE UNTIL PROCESSOR IS UPGRADED TO SYSTEM=(xxxx,n)

PROCESSOR
Explanation: You wrote an IOCDS in preparation of a CPC upgrade using keyword CHECKCPC=NO in the PARM
parameter of the EXEC statement (z/OS or VSE) or using CMS option NOCHKCPC. IOCP generated an IOCDS
using the machine limits and rules for a SYSTEM=(xxxx,n) CPC. However, your current CPC is not associated with
SYSTEM=(xxxx,n). Therefore, the IOCDS cannot be used to perform a power-on reset until your current CPC is
upgraded to a CPC associated with SYSTEM=(xxxx,n).
System action: IOCP prints this caution message after the last IOCP configuration report has been printed. IOCP
continues processing.
Programmer response: If your CPC is scheduled to be upgraded to a SYSTEM=(xxxx,n) CPC, do not use this IOCDS
before the upgrade and inform the service representative which IOCDS to use after the upgrade. If your CPC is not
scheduled to be upgraded, change the SYSTEM keyword value to match your CPC and specify CHECKCPC=YES or
CHECKCPC when running IOCP.
ICP065I **WARNING** THE CONFIGURATION IS UNUSABLE ON THE EC LEVEL OF THIS

PROCESSOR (xxxxxxxx xxxxxxxx)
Explanation: Beginning with ICP IOCP 1.2.0, your configuration is unusable on the EC level of this CPC. The
IOCDS contains an unsupported function or channel path type for this model CPC at its current EC level. xxxxxxxx
xxxxxxxx is a hexadecimal value that indicates the unsupported function or channel type. See Table 26 on page 296 to

ICP066I
determine the meaning of the hexadecimal value. If the hexadecimal value is not defined in this edition of the IOCP
User's Guide, see the latest edition of the document.
For ICP IOCP 1.1.0, this message is a caution message and the following explanations apply:
v For batch IOCP users, your configuration is unusable on some EC levels of this CPC. The IOCDS contains an
unsupported function or channel path type for this model CPC at some ECs. xxxxxxxx xxxxxxxx is a hexadecimal
value which indicates the possible unsupported function or channel type. See Table 26 on page 296 to determine
the meaning of the hexadecimal value. If the hexadecimal value is not defined in this edition of the IOCP User's
Guide, see the latest edition of the document.
To determine what the EC installed on your CPC supports, select the I/O Configuration task on the Support
Element and click Supported I/O mask under View. If the IOCDS contains anything that your CPC does not
support, the display includes Unsupported mask items, which shows the exact items that are not supported.
v For stand-alone IOCP users, your configuration is unusable on the EC level of this CPC. The IOCDS contains an
unsupported function or channel path type for this model CPC at its current EC level. xxxxxxxx xxxxxxxx is a
hexadecimal value which indicates the unsupported function or channel type. See Table 26 on page 296 to
determine the meaning of the hexadecimal value. If the hexadecimal value is not defined in this edition of the
IOCP User's Guide, see the latest edition of the document.
For more information about this message, see “CPC activation and power-on reset error” on page 296.
System action: IOCP prints this message after all reports have been processed and continues processing.
Programmer response: Beginning with ICP IOCP 1.2.0, do not attempt to use the IOCDS at the current EC.
Unsupported functions or channel types prevent an IOCDS from being used for power-on reset. If you intend to use
the IOCDS at the current EC, write a new IOCDS that does not use the unsupported functions or channel path types.
If you intend to use the IOCDS at a new EC following an upgrade and that EC supports the function or channel
type, ignore this message.
For ICP IOCP 1.1.0, the following programmer responses apply:
v For batch IOCP users, if the EC installed on your CPC supports your configuration, ignore this message.
Otherwise, do not attempt to use the IOCDS at the current EC. Unsupported functions or channel types prevent an
IOCDS from being used for power-on reset. If you intend to use the IOCDS at the current EC, write a new IOCDS
that does not use the unsupported functions or channel path types. If you intend to use the IOCDS at a new EC
following an upgrade and that EC supports the function or channel type, ignore this message.
v For stand-alone IOCP users, do not attempt to use the IOCDS at the current EC. Unsupported functions or channel
types prevent an IOCDS from being used for power-on reset. If you intend to use the IOCDS at the current EC,
write a new IOCDS that does not use the unsupported functions or channel path types. If you intend to use the
IOCDS at a new EC following an upgrade and that EC supports the function or channel type, ignore this message.
If this message is in a disassembled I/O configuration source file after your CPC is upgraded, disassemble the
IOCDS again to remove this message.
ICP066I **WARNING** THE IOCDS CONFIGURATION CONTAINS INFORMATION NOT SUPPORTED

BY THIS LEVEL OF IOCP (xxxxxxxx xxxxxxxx)
Explanation: IOCP is processing an IOCDS to print reports for batch or stand-alone, or to disassemble the IOCDS
into an I/O configuration source (IOCS) file on the Support Element hard disk for stand-alone only. The IOCDS was
written by a higher level of IOCP than the one currently processing it and contains a function or channel path type
that the current IOCP level does not support. xxxxxxxx xxxxxxxx is a hexadecimal value that indicates the
unsupported function or channel type. See Table 26 on page 296 to determine the meaning of the hexadecimal value.
If the hexadecimal value is not defined in this edition of the IOCP User's Guide, see the latest edition of the
document.
The current level of IOCP cannot correctly process the unsupported function or channel type.
System action: IOCP prints this message after all reports have been processed and continues processing. For
stand-alone IOCP function Disassemble Data Set, IOCP includes this message at the top of the IOCS file that IOCP
creates on the Support Element hard disk.
Programmer response: Use the same level of IOCP to report on the IOCDS as the level of IOCP that wrote it. Do
not use stand-alone IOCP to process this IOCDS. Do not use a disassembled IOCS containing this message because it
can have incorrect or missing information.
For stand-alone IOCP users, do not attempt to use the IOCDS at the current EC. For batch IOCP users, if your level
of batch IOCP is greater than or equal to the level of stand-alone IOCP installed on this CPC, do not attempt to use

ICP067I • ICP069I
the IOCDS at the current EC. Unsupported functions or channel types prevent an IOCDS from being used for
power-on reset. If you intend to use the IOCDS at the current EC, write a new IOCDS that does not use the
unsupported functions or channel path types. If you intend to use the IOCDS at a new EC following an upgrade and
that EC supports the function or channel type, ignore this message.
ICP067I **WARNING** IOCDS WAS WRITTEN WITH IOCP VERSION xx.xx.xx WHICH IS HIGHER
THAN VERSION OF IOCP PROCESSING IT
Explanation: IOCP is processing an IOCDS to print reports for batch or stand-alone or to disassemble the IOCDS
into an I/O configuration source (IOCS) file on the Support Element hard disk for stand-alone only. The IOCDS was
written by a higher level of IOCP than the one currently processing it, and it might contain something the current
IOCP level does not support, for example, a new keyword.
System action: IOCP prints this message after all reports have been processed and continues processing. For
stand-alone IOCP function Disassemble Data Set, IOCP includes this message at the top of the IOCS file IOCP creates
on the Support Element hard disk.
Programmer response: Use the same level of IOCP to report on the IOCDS as the level of IOCP that wrote it. For
stand-alone IOCP function Disassemble Data Set, use the same level or a higher level of IOCP. Do not disassemble
until after the correct EC is installed. If you receive this message without messages ICP065I and ICP066I, the
disassembled IOCS can be used if no significant differences exist between the IOCP that wrote the IOCDS and the
current IOCP.
ICP068I **WARNING** IOCP WAS UNABLE TO DETERMINE IF THE CONFIGURATION IS USABLE ON

THE EC LEVEL OF THIS PROCESSOR (CODE xxxxxxxx)
Explanation: IOCP attempted to determine the IOCP functions and channel path types supported by the EC level of
your CPC using the I/O supported mask but was unsuccessful. Therefore, IOCP was unable to perform the
verification involved with message ICP065I. Any IOCDS written by IOCP is valid for your model CPC at the latest
EC level. However, IOCP was unable to verify that all of the functions and channel path types defined in the IOCDS
are supported by the specific EC level of your CPC.
System action: IOCP prints this warning message after all reports have been processed and continues processing.
Programmer response: This message results when you are writing an IOCDS from a CPC this version of IOCP does
not support (for example, 2064) to a CPC this program does support (for example, 2094). Ignore it. If you are writing
an IOCDS from a CPC this version of IOCP supports, notify your service representative to determine why IOCP was
unsuccessful in reading the I/O supported mask from your CPC.
Any IOCDS written by IOCP is valid for your model CPC at the latest EC level. If the IOCDS contains any function
or channel path type not supported by the EC level of your CPC, you cannot use the IOCDS for power-on reset until
you upgrade to the appropriate EC level.
Problem determination: If you are running IOCP on VM or as a guest on VM, see “IOCP command – usage notes
for VM” on page 125 for possible reasons for the unsuccessful operation.
ICP069I IOCP ASSIGNED A DEFAULT VALUE TO MAXDEV KEYWORD

Explanation: Beginning with ICP IOCP 1.3.0, IOCP assigned a default value to the MAXDEV keyword. The IOCP
input supports dynamic I/O configuration. If the MAXDEV keyword was not specified, IOCP assigns a default value
for subchannel set 0 for each defined CSS and each subchannel set other than 0 that contains devices. If the
MAXDEV keyword was specified, IOCP does the following actions:
v Assigns a default value for subchannel set 0 for each CSS that was defined on the RESOURCE statement but was
omitted from the MAXDEV keyword.
v Assigns a default value for each subchannel set other than 0 that has devices but was omitted from the MAXDEV
keyword.
The IOCP default value used is the maximum number of devices your CPC supports for a subchannel set. The
hardware system area (HSA) space is the maximum for the subchannel set. For additional information, see Table 18
on page 287. This message is an informational message only.
System action: IOCP prints this informational message at the end of the IOCDS Totals Report and continues
processing.

ICP070I • ICP100I
Programmer response: In the device totals section of the IOCDS Totals Report, determine which subchannel sets
were assigned a default value. Each default value has a special indicator following it (for example, '***'). If you want
to do so, specify a value for each defined subchannel set and CSS in the MAXDEV keyword.
ICP070I SEARCH FOR '*ICP' TO FIND EACH IOCP MESSAGE

Explanation: Stand-alone IOCP performed the action Build Data Set and included this message at the top of the
IOCP source file (IOCS). Search for the character string *ICP to find all IOCP messages in the IOCS.
System action: IOCP includes this informational message at the top of the IOCS when building an IOCDS and
continues processing.
Programmer response: To search for IOCP messages in a source file, search for the characters in the quoted string.
ICP071I IOCP GENERATED A DYNAMIC TOKEN HOWEVER DYNAMIC I/O CHANGES ARE NOT
POSSIBLE WITH THIS IOCDS
Explanation: Your CPC has a fixed storage size for the Hardware System Area (HSA) and therefore requires a
dynamic token in every IOCDS. Either no token was specified in the IOCP source file or a change was made to the
card-image file with the stand-alone IOCP editor. Therefore, IOCP generated a token of its own. However, no
operating system supports dynamic I/O configuration with this IOCDS.
System action: IOCP prints this informational message at the end of the IOCDS Totals Report and continues
processing.
Programmer response: For Hardware Configuration Definition (HCD) users, rebuild the IOCP source file from HCD
and rewrite the IOCDS. For z/VM users who do not use HCD but want dynamic I/O configuration, specify the DYN
option when running IOCP and rewrite the IOCDS. Otherwise, do nothing. For all other operating system users, do
nothing. For stand-alone IOCP, if the IOCDS must support dynamic I/O configuration, import the I/O configuration
source file again and build the data set without using the editor to change anything. If dynamic I/O configuration is
not required, keep the modified IOCDS.
ICP072I **WARNING** IOCP WAS UNABLE TO QUERY THE TARGET CPC (CODE xxxxxxxx)
Explanation: The HCD requested that IOCP write an IOCDS to a target CPC. IOCP performed a query operation of
the target CPC in preparation of the write, and the query failed with the listed code.
System action: IOCP prints this warning message following the ID statement and continues processing.
Programmer response: If an error occurred writing the IOCDS to the target CPC, perform the programmer response
for the listed error message. If the IOCDS was successfully written, notify your service representative to determine
why IOCP was unsuccessful in querying the target CPC in preparation of writing the IOCDS. The IOCDS is usable.
ICP073I IOCP VERSION nn.nn.nnn

Explanation: Stand-alone IOCP writes an I/O configuration source (IOCS) file for the Build data set and
Disassemble data set functions. IOCP always includes this message in the IOCS file. It is an informational message to
indicate the version of IOCP writing the data set.
System action: IOCP writes an IOCS containing this message and continues processing.
Programmer response: No response is required.
ICP100I SYMBOL IN NAME FIELD EXCEEDS 8 CHARACTERS AND/OR CONTAINS INVALID

CHARACTERS
Explanation: The symbolic name (label) in the name field contains more than eight characters or contains incorrect
characters. It must contain 1 - 8 alphanumeric or special (#, @, $) characters. The first character must be alphabetic or
special. Only uppercase characters are allowed.
System action: IOCP continues to validate and check the syntax of the remaining statements. However, it does not
generate an IOCDS and it does not produce configuration reports.
Programmer response: Correct the name field.

ICP101I • ICP107I
ICP101I OPERATION FIELD NOT FOUND

Explanation: The operation field (statement name) was not found before column 72. The statement name must have
one or more blanks preceding and following it. A blank line results in this error.
Programmer response: Check for format errors and correct the errors.
ICP102I UNSUPPORTED OPERATION - xxxxxxxx

Explanation: IOCP did not recognize string xxxxxxxx as a valid IOCP statement. The statement name must have one
or more blanks preceding and following it.
Programmer response: Use a valid IOCP statement name in the operation field.
ICP103I **WARNING** ID STATEMENT IGNORED - OUT OF SEQUENCE

Explanation: IOCP found an ID statement after processing one or more other IOCP statements.
System action: IOCP ignores the ID statement that is out of sequence and continues processing.
Programmer response: Ensure that the ID statement is the first IOCP statement in the input.
ICP104I **WARNING** MORE THAN ONE ID STATEMENT - ONLY THE FIRST WAS USED
Explanation: More than one ID statement was found in the input statements. Only one ID statement can be
specified.
System action: IOCP processes the first ID statement, ignores any following ID statements, and continues
processing.
Programmer response: Ensure that there is only one ID statement in the input and that it is the first IOCP
statement.
ICP105I OPERAND FIELD NOT FOUND

Explanation: The operand field was not found before column 72.
Programmer response: Correct the operand field.
ICP106I NON-BLANK CHARACTER(S) FOUND IN COLUMNS 1-15 OF CONTINUATION CARD

Explanation: A nonblank character was found in columns 1 - 15. On a continuation card, columns 1 - 15 must be
blank, and the operand field must start in column 16. (The preceding card had a nonblank character in column 72,
indicating that this card is a continuation card.)
Programmer response: Ensure that the operand field on the continuation card starts in column 16.
ICP107I **WARNING** EXCEEDED THE ALLOWED NUMBER OF COMMENT CONTINUATION CARDS

Explanation: More than two continuation cards were found for an IOCP comment card.
System action: IOCP prints the extra comment continuation cards and continues processing.
Programmer response: Restructure the comment so that it can be contained in a total of three comment cards, or
code an asterisk in column 1 and remove the continuation characters in column 72 of the preceding cards.

ICP108I • ICP113I
ICP108I UNEXPECTED END OF FILE ON SYSTEM INPUT

Explanation: A continuation card was expected when an end-of-file condition occurred on the input data set. The
last card contains a nonblank character in column 72.
System action: IOCP does not generate an IOCDS and does not produce configuration reports.
Programmer response: Ensure that no statements or comment cards are missing in the input.
ICP109I **WARNING** INVALID CHARACTER(S) IN MSG1 | MSG2 STRING TRANSLATED TO A

VALID CHARACTER
Explanation: The message string for keyword MSG1 or MSG2 in the ID statement contains one or more incorrect
characters. Each incorrect character is translated to a double quotation mark (") and is shown as such in the
configuration report headings and on the IOCDSM frame of the Support Element. See the “English (U.S.) I/O
Interface Code” table for EBCDIC characters in the 3270 IDS Character Set Reference, GA27-2837, for a list of valid
printable characters.
System action: IOCP translates each incorrect character to the valid character and continues processing.
Programmer response: Correct the message string for MSG1 or MSG2.
ICP110I EXTRA LEFT PARENTHESIS - xxxxxxxx

Explanation: IOCP detected an extra left parenthesis as the first character of string xxxxxxxx. The parameter requires
one or two left parentheses to start the keyword value field.
Programmer response: Correct the parameter.
ICP111I DELIMITER ERROR, EXPECT COMMA - xxxxxxxx

Explanation: IOCP expected a comma as the first character of string xxxxxxxx. The parameter requires that you use
a comma to separate the values.
Programmer response: Correct the parameter or the delimiter.
ICP112I DELIMITER ERROR, EXPECT BLANK OR COMMA - xxxxxxxx

Explanation: IOCP expected a blank or comma as the first character of string xxxxxxxx. This error can occur if a
keyword value contains too many characters.
ICP113I DELIMITER ERROR, EXPECT COMMA OR RIGHT PARENTHESIS - xxxxxxxx

Explanation: IOCP expected a comma or right parenthesis as the first character of string xxxxxxxx. This error can
occur if a keyword value contains too many characters.

ICP114I • ICP118I
ICP114I DELIMITER ERROR, EXPECT LEFT PARENTHESIS - xxxxxxxx

Explanation: IOCP expected a left parenthesis as the first character of string xxxxxxxx. The parameter requires that
you enclose the keyword value in parentheses.
Programmer response: Check for incorrect characters in the parameter and correct the delimiter.
ICP115I DELIMITER ERROR, EXPECT RIGHT PARENTHESIS - xxxxxxxx

Explanation: IOCP expected a right parenthesis as the first character of string xxxxxxxx. This error can occur if a
keyword value contains too many characters or if more parameters than your CPC supports were specified.
Programmer response: Correct the parameter and correct the incorrect delimiter with a right parenthesis.
ICP116I **WARNING** CONTINUATION CARD TREATED AS A COMMENT BY IOCP

Explanation: IOCP issues this message following a continuation card that is treated as a comment by IOCP (that is,
it is ignored). Any continuation cards for the same statement that follow this message are also treated as comments
by IOCP.
One of the following conditions is true:
v Card n is continued but the operand field on card n ends with a blank instead of a comma. Card n+1 receives this
message and any keywords on card n+1 are ignored.
v Card n is continued but the operand field on card n+1 begins in column 17 or greater instead of column 16. Card
n+1 receives this message and any keywords on card n+1 are ignored.
System action: IOCP ignores the continuation cards being treated as comments and continues processing.
Programmer response: Verify that there are no keywords on the continuation cards being treated as comments. If
there are, fix the conditions that caused IOCP to treat them as comments. If there are no keywords and a valid
comment exists on the continuation card, ignore the message or move the comment to a comment card. Comment
cards begin with an asterisk (*) in column 1 or a period and asterisk (.*) in columns 1 and 2. See Appendix A,
“Coding IOCP statements,” on page 237 for more information about how to code IOCP statements and comments.
ICP117I INVALID CONTROL UNIT NUMBER

Explanation: An incorrect control unit number was entered as the value for keyword CUNUMBR on either a
CNTLUNIT or an IODEVICE statement. Control unit numbers must be one to four hexadecimal digits in the range
0000-FFFE. FFFF is not a valid control unit number.
Programmer response: Correct the control unit number.
ICP118I PARAMETER CONTAINS INVALID CHARACTERS

Explanation: A parameter contains one or more incorrect characters. See the “English (U.S.) I/O Interface Code”
table for EBCDIC characters in 3270 IDS Character Set Reference GA27-2837, for a list of valid printable characters.
Programmer response: Correct the incorrect characters in the parameter.

ICP119I • ICP124I
ICP119I BEGINNING APOSTROPHE NOT FOUND

Explanation: The beginning apostrophe was not found in a parameter that requires the keyword value to be
enclosed in apostrophes.
ICP120I ENDING APOSTROPHE NOT FOUND WITHIN ALLOWED LENGTH

Explanation: The ending apostrophe was not found in a parameter that requires the keyword value to be enclosed
in apostrophes.
ICP121I OPERAND FIELD IS TOO LONG TO PROCESS

Explanation: The length of the operand field is greater than the length IOCP can process for operand fields.
Maximum length for an operand field is 1024 characters.
Programmer response: Check that the statement and continuation cards are correctly specified.
ICP122I REQUIRED KEYWORD IS MISSING - xxxxxxxxx

Explanation: One or more required keywords were not specified in the statement. String xxxxxxxxx is one of the
required keywords that is missing. String xxxxxxxxx is blank if no keyword is specified on the ID statement. The ID
statement does not require a specific keyword but requires specifying at least one keyword.
The SYSTEM keyword on the ID statement is required when IOCP is writing an IOCDS in preparation of a CPC
upgrade (that is, keyword CHECKCPC=NO is specified in the PARM parameter of the EXEC statement (z/OS or
VSE) or CMS option NOCHKCPC is specified).
Programmer response: Ensure that all required keywords are specified.
ICP123I DUPLICATE KEYWORDS ON STATEMENT

Explanation: A keyword parameter was found more than one time in the statement. Keyword parameters cannot be
repeated in the same statement.
Programmer response: Ensure that keyword parameters are not duplicated.
ICP124I MISSING A REQUIRED KEYWORD FOR UNSPANNED CHPID

Explanation: The NOTPART, PARTITION, SHARED, or IOCLUSTER keyword must be specified for all unspanned
channel paths. An unspanned channel path is defined to only one CSS.
Programmer response: Ensure that each CHPID statement defining an unspanned channel path specifies the
NOTPART, PARTITION, SHARED, or IOCLUSTER keyword.

ICP125I • ICP130I
ICP125I PARAMETER CONTAINS NON-NUMERIC CHARACTERS - xxxxxxxx

Explanation: In string xxxxxxxx, a character other than 0-9 was found in a parameter for a keyword value that
requires a decimal value.
Programmer response: Correct the keyword value.
ICP126I PARAMETER CONTAINS NON-HEXADECIMAL CHARACTERS OR INCORRECT LENGTH -

xxxxxxxx
Explanation: In string xxxxxxxx, a character other than 0-9 or A-F was found in a parameter for a keyword value
that requires a hexadecimal value, or not enough digits were specified in a keyword value. (The channel path
identifier in the PATH and CPATH parameters requires two digits. Also, link addresses in the LINK parameter
require two or four digits.)
ICP127I UNKNOWN KEYWORD - xxxxxxxx

Explanation: IOCP found an unknown keyword parameter in string xxxxxxxx.
Programmer response: Correct the unknown keyword parameter.
ICP128I INVALID PROTOCL VALUE. MUST BE D, S, OR S4

Explanation: A character other than D, S, or S4 was found as the keyword value in the PROTOCL parameter.
ICP129I INVALID SHARED VALUE. MUST BE Y OR N

Explanation: A character other than Y or N was found as the keyword value in the SHARED parameter of the
CNTLUNIT statement.
ICP130I **WARNING** [IOCDS WILL NOT SUPPORT DYNAMIC I/O CONFIGURATION | DYNAMIC
TOKEN ALTERED]
Explanation: A change was made to the card-image file with the stand-alone IOCP editor. For CPCs that do not
have a fixed storage size for the Hardware System Area (HSA), dynamic I/O configuration is not supported by the
IOCDS written from the modified card-image file and the following conditions exist:
v The MAXDEV keyword on the RESOURCE statement is ignored
v Managed channel paths are unusable
v Any reserved logical partitions existing in the hardware system area (HSA) space cannot be used.
For CPCs that have fixed HSA, IOCP altered the dynamic token and dynamic I/O configuration is not supported by
any operating system. See Table 16 on page 285 to determine which CPCs have fixed HSA.
System action: IOCP prints this warning message following the ID statement and continues processing.

ICP131I • ICP136I
Programmer response: If the IOCDS must support dynamic I/O configuration, import the I/O configuration source
(IOCS) file again and build the data set without using the editor to change anything. If dynamic I/O configuration is
not required, keep the modified IOCDS.
ICP131I CONTINUATION CHARACTER EXPECTED BUT NOT FOUND

Explanation: IOCP expected to find a continuation character (any nonblank character in column 72) and did not
find one. IOCP expects a continuation character when the operand field of a statement ends with a comma. IOCP
also expects a continuation character when the operand field of a statement ends in column 71 and the next card has
blanks in columns 1-15 and a nonblank in column 16.
Programmer response: If you want a continued statement, place a nonblank character in column 72. If you do not
want a continued statement, remove the condition that caused IOCP to expect a continuation character (for example,
remove the trailing comma).
ICP132I INVALID TIMEOUT VALUE. MUST BE Y OR N

Explanation: A character other than Y or N was found as the keyword value in the TIMEOUT parameter.
ICP133I INVALID TYPE VALUE

Explanation: The TYPE keyword on the CHPID or FUNCTION statement specified an invalid value. For the CHPID
statement, see “CHPID” on page 63 for the valid TYPE keyword values. For the FUNCTION statement, see
“FUNCTION” on page 103 for the valid TYPE keyword values.
ICP134I UNIT OR MODEL PARAMETER CONTAINS INVALID CHARACTERS

Explanation: An incorrect character was found in the UNIT or MODEL parameter. The allowable characters are
alphanumeric and special (#, @, and $). Only uppercase characters are allowed. The UNIT keyword on the
CNTLUNIT statement also accepts a hyphen (-) as a valid character.
Programmer response: Correct the UNIT or MODEL parameters.
ICP135I INVALID UNITADD NUMBER

Explanation: In the UNITADD parameter, a keyword value for number was not within the allowable range of
decimal 1-256.
ICP136I INVALID DEVICE NUMBER COUNT

Explanation: In the ADDRESS parameter, a keyword value for count was not within the allowable range of decimal
1-256.

ICP137I • ICP140I
ICP137I CHPID IS NOT SHARED AND NOT RECONFIGURABLE - INVALID CANDIDATE LIST
Explanation: The PART or PARTITION keyword has an incorrect candidate list because the channel path is not
shared or reconfigurable. The following conditions in the PART or PARTITION keyword produce this error for
unshared and non-reconfigurable channel paths:
v Null access list with no candidate list
v Null access list with a candidate list of two or more logical partitions (LP)
v An access list of one LP with a candidate list specifying a different LP.
Programmer response: For a reconfigurable channel path, specify the REC parameter in the PART or PARTITION
keyword. For a shared channel path, specify the SHARED keyword or replace the PART or PARTITION keyword
with the NOTPART keyword. Otherwise, do one of the following actions:
v For a null access list, specify a candidate list with one LP
v For an access list of one LP, delete the candidate list or change it to an equal sign (=).
ICP138I NULL CANDIDATE LIST SPECIFIED FOR CHPID xx IN CSS y

Explanation: You generated a candidate list that contained no logical partitions (a null candidate list) for channel
path xx in CSS y by specifying in the CHPID statement one of the following groups of values:
v A zero as the access list and an equal sign (=) as the candidate list for CSS y using the PART, PARTITION, or
NOTPART keyword.
v All logical partitions (LPs) in the configuration for CSS y in the candidate list parameter of the NOTPART
keyword.
A channel path must have at least one LP in its generated candidate list from each CSS in which it is defined.
Programmer response: In the CHPID statement, do one of the following actions:
v Specify at least one LP in the candidate list for each CSS
v Delete at least one LP from CSS y in the candidate list parameter of the NOTPART keyword
v For a spanned channel path that you do not want to access any LP in a CSS, remove the CSS ID from the PATH
keyword of the CHPID statement.
ICP139I **WARNING** PCHID KEYWORD WAS NOT SPECIFIED ON EVERY CHPID STATEMENT
REQUIRING IT
Explanation: IOCP found one or more CHPID statements for a deck verification where the PCHID keyword is
required but was not specified. IOCP performs a deck verification if the WRTCDS PARM keyword is not specified or
if WRTCDS=NO or NOWRTCDS is specified. The PCHID keyword is required for all channel paths except ICP, CIB,
CS5, and IQD (TYPE keyword).
System action: IOCP prints this message after the last statement is processed and then continues processing the
configuration reports.
Programmer response: Specify the PCHID keyword on every CHPID statement that requires it before attempting to
write an IOCDS. The CHPID mapping tool can be used to assign PCHID numbers to channel paths. See “Using a
PCHID report to help plan your I/O configuration” on page 17.
ICP140I INVALID xxxxxxx VALUE

Explanation: The value specified for keyword xxxxxxx does not have the correct type of characters or is not
supported by your CPC. See the following explanations:

ICP141I • ICP143I
v For a PCHID keyword value, each character must be hexadecimal (0-9 or A-F), and your CPC must support the
number.
v For a VCHID keyword value, each character must be hexadecimal (0-9 or A-F), and your CPC must support the
number.
v For an AID keyword value, each character must be hexadecimal (0-9 or A-F), and your CPC must support the
number.
v For a PORT keyword value, each character must be decimal (0-9), and your CPC must support the number.
v For a CSYSTEM keyword value, each character must be alphanumeric (0-9 or A-Z).
v For an LSYSTEM keyword value, each character must be alphanumeric (0-9 or A-Z).
v For a PNETID keyword value, each character must be alphanumeric (0-9 or A-Z).
v For a FID keyword value, each character must be hexadecimal (0-9 or A-F), and your CPC must support the
number.
v For a UID keyword value, each character must be hexadecimal (0-9 or A-F) and cannot be zero.
v For a VF keyword value, each character must be decimal (0-9) and in the valid range for the given function type.
See Table 18 on page 287 to determine what your CPC supports. See Table 10 on page 106 to determine the valid VF
range for each function type.
generate an IOCDS and it does not produce configuration reports
ICP141I **WARNING** MORE THAN ONE [RESOURCE | UUID] STATEMENT - ONLY THE FIRST WAS
USED
Explanation: More than one RESOURCE or UUID statement was found in the input statements. Only one
RESOURCE or UUID statement can be specified.
System action: IOCP processes the first RESOURCE or UUID statement, ignores any following RESOURCE or UUID
statements, and continues processing.
Programmer response: Ensure the following:
v That there is only one RESOURCE statement in the input and that it follows the ID statement and precedes all
CHPID statements.
v That there is only one UUID statement in the input and that it follows the RESOURCE statement and precedes all
FUNCTION statements.
ICP142I RESOURCE STATEMENT IS REQUIRED

Explanation: IOCP did not find a RESOURCE statement or the RESOURCE statement contained errors. The
RESOURCE statement is required.
Programmer response: Ensure that the RESOURCE statement is specified, that it precedes all CHPID statements,
and it contains no errors.
ICP143I CHPID ERROR. CODE=xx

Explanation: IOCP detected one of the following errors on the CHPID statement:
Code Description
01 The IOCLUSTER keyword value is incorrect. The value must contain a maximum of eight alphanumeric or
special (@, #, or $) characters.
02 Too many I/O cluster names have been defined with the IOCLUSTER keyword. See Appendix F, “Machine
limits and rules,” on page 285 to determine the maximum number of I/O clusters your CPC supports..
03 The IOCLUSTER keyword and PART, PARTITION, or NOTPART keyword were specified on the same
CHPID statement. Specify only one of these keywords.

ICP144I • ICP147I
04 The IOCLUSTER keyword was specified for a channel path that is not managed (CHPARM keyword). The
IOCLUSTER keyword is valid only for managed channel paths.
05 A channel path cannot be specified as spanned (defined in more than one CSS) and managed (CHPARM
keyword).
06 Managed channel paths require the IOCDS to be dynamic-capable (TOK keyword). A dynamic-capable
IOCDS is built when using HCD or specifying IOCP CMS utility option DYN for VM.
Programmer response: Ensure that the CHPID statement is coded correctly.
ICP144I CNTLUNIT ERROR. CODE=xx

Explanation: IOCP detected one of the following errors on the CNTLUNIT statement:
Code Description
01 Only pairs of asterisks (**) were specified in the PATH keyword parameter. The PATH keyword parameter
must contain at least one chpid number for each specified CSS.
02 All the channel paths attaching to a managed control unit (PATH keyword) must be shared.
03 The control unit is managed (PATH keyword) but a channel path type other than CNC, FCV, or FC attaches
to the control unit. Only CNC, FCV, and FC channel paths can attach to a managed control unit.
04 A link address (LINK keyword) other than two asterisks (**) was specified for a channel path of two
asterisks (**) in the PATH keyword on the CNTLUNIT statement. The link address must be two asterisks.
05 A managed channel path (CHPARM keyword on CHPID statement) is defined on a control unit. Managed
channel paths cannot be defined on a control unit.
Programmer response: Ensure that the CNTLUNIT statement is coded correctly.
ICP145I DEVICE NUMBER EXCEEDS HEX FFFF

Explanation: In the ADDRESS parameter of the IODEVICE statement, the keyword value for the device number
plus the value for count minus one, exceeds the allowable value of hexadecimal FFFF for a device number.
Programmer response: Correct the keyword values.
ICP146I INVALID STADET VALUE. MUST BE Y OR N

Explanation: A character other than Y or N was found as the keyword value in the STADET parameter.
ICP147I INVALID PARTITION NAME - xxxxxxxx

Explanation: The PARTITION name xxxxxxxx contains more than eight characters or contains incorrect characters.
The PARTITION name must contain from one to eight alphanumeric characters, the first of which must be alphabetic.
Only uppercase characters are allowed. You can also specify an asterisk (*) to reserve a logical partition until a future
dynamic I/O configuration change is made to assign a name to the logical partition. The words PHYSICAL, REC,
SYSTEM, and PRIMxxxx (where xxxx is a 4-digit decimal number) are reserved and cannot be used as the names of
logical partitions.
You cannot specify a logical partition name of zero (0) for a candidate list on a CHPID statement. If you specify zero

ICP148I • ICP152I
(0) for a null channel path access list or a null device candidate list, you cannot also specify a logical partition name.
ICP148I INVALID PARTITION PARAMETER - xxxxxxxx

Explanation: The PARTITION keyword has an incorrect value beginning with the first character of string xxxxxxxx.
IOCP expected the REC parameter value.
ICP149I PARTITION xxxxxxxx HAS NOT BEEN DEFINED [FOR CSS y]

Explanation: You specified partition name xxxxxxxx, which was not defined by the RESOURCE statement or was
defined but not for CSS y. The RESOURCE statement must define all partition names in the configuration. Also, the
CSS ID number, if specified, must be the one to which the partition has access.
Programmer response: Ensure the partition name on the CHPID or IODEVICE statement is correct. If the partition
name is correct but undefined, add it to the RESOURCE statement. If the partition name is correct but the CSS ID
number is not, correct the CSS ID number.
ICP150I INVALID SYSTEM PARAMETER

Explanation: The SYSTEM parameter contains an incorrect value. The valid values are listed in Appendix F,
“Machine limits and rules,” on page 285.
Programmer response: Correct the SYSTEM keyword value or remove the keyword.
ICP151I SYSTEM KEYWORD IGNORED. SYSTEM=(xxxx,n) USED BY ICP IOCP

Explanation: The SYSTEM keyword is ignored by the stand-alone version of IOCP and also when IOCP is writing
an IOCDS. IOCP has used the machine limits and rules associated with the parameter SYSTEM=(xxxx,n). You have
specified a SYSTEM keyword value that does not support your CPC. For additional information, see Appendix F,
“Machine limits and rules,” on page 285. This message is an informational message only.
System action: IOCP prints this message after the ID statement and then continues processing.
Programmer response: For stand-alone IOCP or when writing an IOCDS to a CPC that is not being upgraded,
ignore the message or change the SYSTEM keyword value to match your CPC. If you are upgrading your CPC to a
CPC that requires a different SYSTEM keyword value, you can do one of the following actions:
v Use the SYSTEM keyword value to verify that the IOCP input file is valid (IOCP processing parameter
WRTCDS=NO or NOWRTCDS). You can give this IOCP input file to the service representative who will upgrade
the CPC.
v Use the SYSTEM keyword value and write the IOCDS to your current CPC (IOCP processing parameter
CHECKCPC=NO or NOCHKCPC) in preparation for its being upgraded. See “Machine rules” on page 285 to
determine if ICP IOCP supports writing an IOCDS to your current CPC in preparation for an upgrade.
ICP152I SYSTEM=(xxxx,n) USED BY ICP IOCP

Explanation: You did not specify the SYSTEM keyword in the ID statement. IOCP sends this message to inform you
that IOCP used the machine limits and rules associated with the parameter SYSTEM=(xxxx,n). This message is an
informational message only.

ICP153I • ICP156I
System action: If an ID statement is present, IOCP prints this message after processing it. If it is not present, IOCP
prints this message after the last statement is processed and then continues processing.
Programmer response: Ignore this message in the following situations:
v IOCP used the correct machine limits and rules for your configuration for a deck verification (WRTCDS=NO or
NOWRTCDS)
v You are writing an IOCDS
v You are using a stand-alone version of IOCP.
Otherwise, specify the correct SYSTEM keyword value on the ID statement. For additional information, see “Machine
You can prevent IOCP from issuing this message by specifying SYSTEM=(xxxx,n) in the ID statement.
ICP153I NOTPART AND PARTITION KEYWORDS CANNOT SPECIFY THE SAME CSS
Explanation: You cannot specify the same CSS ID in both the NOTPART and PARTITION keywords on the same
CHPID or IODEVICE statement. Specify the CSS ID in only one of these keywords on a statement.
Programmer response: Remove the duplicate CSS ID from the NOTPART or PART or PARTITION keyword from
the statement.
ICP154I INVALID MIF IMAGE ID NUMBER

Explanation: The PARTITION parameter on the RESOURCE statement contains an incorrect MIF image ID number.
The number must be a 1-digit hexadecimal number in the range of 1-F.
ICP155I DUPLICATE MIF IMAGE ID NUMBERS SPECIFIED

Explanation: You specified the same MIF image ID number for more than one partition in a CSS on the RESOURCE
statement. Every partition in a CSS must have a unique MIF image ID number.
ICP156I DUPLICATE PARTITION NAMES SPECIFIED

Explanation: You specified the same partition name more than one time on the RESOURCE statement, an
IODEVICE statement, or in the access list or the candidate list on a CHPID statement. You are not permitted to
duplicate partition names in the following:
v RESOURCE statement
v Access list of a CHPID statement
v Candidate list of a CHPID statement
v IODEVICE statement
System action: For the RESOURCE statement, IOCP ends with a return code of 8. For all other statements, IOCP
continues to validate and check the syntax of all other statements. It does not generate an IOCDS and it does not
produce configuration reports.

ICP157I • ICP160I
ICP157I TOO MANY PARTITION NAMES SPECIFIED

Explanation: You specified too many partition names on the RESOURCE statement, an IODEVICE statement, the
UUID statement, a FUNCTION statement, or in the access list or the candidate list on a CHPID statement. You
cannot specify more than 15 partition names in a CSS, and you cannot specify more partition names than your CPC
supports. For additional information, see “Machine limits” on page 286. You can specify only the maximum number
of partitions names that your CPC supports in the following:
v RESOURCE statement
v Access list of a CHPID statement
v Candidate list of a CHPID statement
v IODEVICE statement
v UUID statement
v FUNCTION statement.
System action: For the RESOURCE statement, IOCP ends with a return code of 8. For all other statements, IOCP
continues to validate and check the syntax of all other statements. It does not generate an IOCDS and it does not
produce configuration reports.
ICP158I CHPIDS CANNOT BE SHARED AND RECONFIGURABLE

Explanation: You specified a channel path as both shared and reconfigurable, which is not allowed. A channel path
is shared if you specify the channel path as spanned or the SHARED keyword or an access list of more than one
logical partition in the CHPID statement. A channel path is reconfigurable if you specify the REC parameter of the
PART or PARTITION keyword in the CHPID statement.
Programmer response: For a shared channel path, delete the REC parameter from the PARTITION keyword.
Otherwise, delete the SHARED keyword or reduce the number of partitions in the access list to one or specify a null
access list.
ICP159I [MAXDEV KEYWORD | RESERVED LP (*)] NOT ALLOWED BECAUSE THE I/O DEFINITION
DOES NOT SUPPORT DYNAMIC I/O CONFIGURATION
Explanation: The MAXDEV keyword or a reserved logical partition (*) was specified for a non-dynamic-capable I/O
configuration (that is, TOK keyword on the ID statement is not present). You can specify the MAXDEV keyword and
reserved logical partitions only for a dynamic-capable IOCDS. A dynamic-capable IOCDS is built when using HCD
or specifying IOCP CMS utility option DYN for VM.
Programmer response: If the IOCDS must support a dynamic-capable I/O configuration, use HCD or IOCP CMS
utility option DYN to write the IOCDS. If dynamic I/O configuration is not required (non-HCD users only), remove
the MAXDEV keyword from the RESOURCE statement and remove all reserved logical partitions (*) from the PART
or PARTITION keyword on the RESOURCE statement.
ICP160I INVALID MAXDEV VALUE - xxxxx

Explanation: An incorrect maximum number of devices was entered as the value for keyword MAXDEV on the
RESOURCE statement. The maximum number of devices can be specified as one to five decimal digits in the range 0
- n, where n is the maximum number of devices your CPC supports. For additional information, see “Machine limits”
on page 286. The maximum number of devices can also be specified as a K value, for example, 63K. A decimal
number must precede the K.

ICP161I • ICP165I
ICP161I **WARNING** MAXDEV VALUE FOR SUBCHANNEL SET y IN CSS x EXCEEDS MAXIMUM
NUMBER ALLOWED
Explanation: The maximum number of devices specified for subchannel set y in CSS x exceeded the maximum
number of devices your CPC supports. For additional information, see “Machine limits” on page 286. IOCP used the
maximum number of devices your CPC supports for the subchannel set.
System action: IOCP prints this message after the RESOURCE statement and then continues processing.
ICP162I **WARNING** MAXDEV KEYWORD NOT SPECIFIED. HSA FOR MAXIMUM NUMBER OF
DEVICES WILL BE ALLOCATED
Explanation: The MAXDEV keyword was not specified on the RESOURCE statement and the IOCP input supports
dynamic I/O configuration. IOCP used the maximum number of devices your CPC supports for subchannel set 0 for
each CSS. IOCP also used the maximum number of devices your CPC supports for subchannel sets other than 0 for
each CSS that has devices defined in that subchannel. The hardware system area (HSA) space is the maximum for
subchannel set 0 for each CSS and for each subchannel set other than 0 that has devices. For additional information,
see “Machine limits” on page 286.
Programmer response: Specify the MAXDEV keyword with the appropriate value or values for each defined CSS.
ICP163I **WARNING** MAXDEV KEYWORD IGNORED

Explanation: The MAXDEV keyword was specified for a non-dynamic-capable I/O configuration that was at some
time dynamic-capable or for a CPC that has fixed HSA. A dynamic-capable IOCDS is built when using HCD on
OS/390, z/OS, or z/VM or specifying IOCP CMS utility option DYN for VM. A dynamic-capable I/O configuration
can be rendered nondynamic when the IOCP input file is modified with stand-alone IOCP or the TOK keyword has
an incorrect value.
For CPCs that have fixed HSA, IOCP ignores the MAXDEV keyword. The maximum number of devices that are
supported by the CPC are available. See Table 16 on page 285 to determine which CPCs have fixed HSA.
System action: IOCP prints this message after the RESOURCE statement, sets the maximum number of devices for
each defined CSS to the maximum number your CPC supports, and then continues processing. See “Machine limits”
on page 286.
Programmer response: For CPCs that do not have fixed HSA, use HCD or IOCP CMS utility option DYN to
regenerate the IOCP input as dynamic. For CPCs that have fixed HSA, remove the MAXDEV keyword.
ICP164I MAXDEV KEYWORD DID NOT SPECIFY A VALUE FOR EVERY DEFINED CSS
Explanation: For ICP IOCP 1.1.0 and 1.2.0, the MAXDEV keyword did not specify a value for each CSS that was
defined on the RESOURCE statement. IOCP used the maximum number of devices your CPC supports for each
unspecified CSS. The hardware system area (HSA) space is the maximum for each unspecified CSS. For additional
information, see “Machine limits” on page 286. This message is an informational message only.
Programmer response: Specify a value for each defined CSS in the MAXDEV keyword.
ICP165I CSS PARAMETER IS REQUIRED ON KEYWORD xxxxxxxx

Explanation: The CSS parameter identifies the CSS to which the xxxxxxxx keyword values apply. CSS is required for
the following keywords and conditions:
v PART or PARTITION on the RESOURCE statement when a CSS ID other than zero is to be defined.
v MAXDEV when more than one CSS is defined.
v PATH on the CHPID and CNTLUNIT statements when more than one CSS is defined.
v NOTPART and PART or PARTITION on the CHPID statement when the channel path is spanned.

ICP166I • ICP170I
v CPATH when the connecting channel path is in a different CSS from the channel path being defined or the channel
path being defined is spanned.
v LINK when the control unit has paths in multiple CSSs.
v PATH, NOTPART, and PART or PARTITION on the IODEVICE statement when the device has access to channel
paths in multiple CSSs.
Programmer response: Specify the CSS parameter on keyword xxxxxxxx.
ICP166I CSS x HAS NOT BEEN DEFINED

Explanation: CSS x is specified in the CSS parameter value but is not defined on the RESOURCE statement.
When connecting CIB, CL5, or CS5 channel paths (TYPE keyword) on the same CPC (that is, internal coupling
connection), the CSS IDs in the CPATH keyword must have been defined on the RESOURCE statement.
Programmer response: If you are defining a CIB, CL5, or CS5 channel path and want an external coupling
connection, change the CSYSTEM keyword value so that it matches the local name (LSYSTEM keyword value on the
ID statement) of the target CPC. For all other definitions, correct the parameter value or define the CSS on the
RESOURCE statement.
ICP167I INVALID CSS ID NUMBER - x

Explanation: An incorrect CSS ID number was specified as the value for the CSS parameter. CSS ID numbers must
be one decimal digit in the range 0 to the maximum your CPC supports. For additional information, see “Machine
Programmer response: Correct the CSS ID number.
ICP168I DUPLICATE CSS ID NUMBERS SPECIFIED

Explanation: A CSS ID number was found more than one time in the CSS parameters of a keyword. CSS ID
numbers cannot be repeated in the CSS parameters of a keyword.
Programmer response: Ensure that CSS ID numbers are not duplicated.
ICP169I CSS x NOT SPECIFIED IN PATH KEYWORD

Explanation: You specified CSS x which was not specified in the CSS parameter of the PATH keyword. On the
CHPID and CNTLUNIT statements, only the CSS ID numbers specified on the PATH keyword can be specified in the
CSS parameters of the other keywords on that statement.
Programmer response: Ensure only CSS ID numbers specified on the PATH keyword for CHPID and CNTLUNIT
statements are specified on other keywords.
ICP170I PARTITION AND NOTPART KEYWORDS SPECIFIED TOO MANY CSS PARAMETERS
Explanation: The total number of CSS ID numbers specified in the CSS parameters of both the PARTITION and
NOTPART keywords exceeds the maximum number of CSSs allowed for your CPC. You cannot specify more CSSs
than your CPC supports. For additional information, see “Machine limits” on page 286.

ICP171I • ICP176I

Programmer response: Reduce the number of CSSs specified by the PARTITION and NOTPART keywords.
ICP171I **WARNING** OS KEYWORD CONVERTED TO CHPARM KEYWORD

Explanation: The OS keyword has been converted to the CHPARM keyword.
Note: Beginning with ICP IOCP 2.1.0, the OS keyword is disallowed.

System action: IOCP prints this warning message following the CHPID statement in question and continues
processing.
Programmer response: Change all OS keywords to the CHPARM keyword.
ICP172I MIF IMAGE ID IS REQUIRED FOR RESERVED PARTITION

Explanation: You specified a reserved logical partition (*) without a MIF image ID. Each reserved logical partition
requires a MIF image ID.
Programmer response: Specify a MIF image ID for every reserved logical partition.
ICP173I PARTITION NAME IS REQUIRED

Explanation: You specified only reserved logical partitions (*) and did not specify any logical partitions with names.
An I/O configuration must have at least one logical partition defined with a name.
Programmer response: Specify a logical partition with a name.
ICP174I NO PARTITION NAMES DEFINED IN CSS x

Explanation: You specified CSS x in the PATH keyword but CSS x contains only reserved logical partitions (*). You
cannot assign a channel path to a CSS that only contains reserved logical partitions.
Programmer response: Assign the channel path to a CSS with a named logical partition.
ICP175I INVALID SCHSET ID NUMBER

Explanation: An incorrect subchannel set ID number was specified as the value for the SCHSET keyword parameter.
Subchannel set ID numbers must be one digit in the range 0 to the maximum your CPC supports. For additional
information, see Table 18 on page 287.
Programmer response: Correct the subchannel set ID number.
ICP176I **WARNING** TOO MANY PARTITION NAMES SPECIFIED FOR xxxx-yyy

Explanation: You specified too many partition names on the RESOURCE statement for a CPC with machine type
xxxx and model number yyy. The machine limits that IOCP used (SYSTEM= keyword) allow more partition names
than model yyy supports. You can power-on reset an IOCDS containing more partition names than model yyy
supports but you can activate only up to the maximum number supported. For additional information, see
Appendix F, “Machine limits and rules,” on page 285.
System action: IOCP prints this message and then continues processing.
Programmer response: If you plan on upgrading to a model that supports more partition names than model yyy, no

ICP177I • ICP182I
change is necessary. If you do not plan on upgrading, reduce the number of partition names to the maximum
supported by model yyy.
ICP177I **WARNING** xxxxxxx KEYWORD IGNORED FOR THIS PROCESSOR

Explanation: You specified keyword xxxxxxx and IOCP ignores that keyword for this CPC. The following keywords
are ignored on some CPCs:
v MIXTYPE keyword on CHPID statement for CPCs prior to 3906
v PORT keyword on FUNCTION statement for 2828 and 2827 CPCs
v TYPE keyword on FUNCTION statement for 2828 and 2827 CPCs.
System action: IOCP prints this and continues processing.
Programmer response: Delete the unsupported keyword from all CHPID or FUNCTION statements.
ICP179I [xxxxxxxx STATEMENT |nnnnn KEYWORD] INVALID FOR THIS PROCESSOR

Explanation: You specified statement xxxxxxxx or keyword nnnnn which is invalid for this CPC. The following are
invalid on some CPCs:
v FUNCTION statement
v PNETID keyword on CHPID statement
v UUID statement
v UID keyword on FUNCTION statement
v VCHID keyword on CHPID statement or FUNCTION statement.
See Table 25 on page 295 to determine which machine types support the statement or keyword in error.
generate an IOCDS and it doe snot produce configuration reports.
Programmer response: Delete the unsupported statement or keyword.
ICP180I NULL CANDIDATE LIST SPECIFIED FOR FID xxxx

Explanation: You generated a candidate list that contained no logical partitions (a null candidate list) for FID xxxx
by specifying in the FUNCTION statement a zero as the access list logical partition (LP) name and an equal sign (=)
as the candidate list using the PART keyword. A function must have at least one LP in its generated candidate list.
Programmer response: In the PART keyword of the FUNCTION statement, do one of the following actions:
v Specify an LP in the access LP name parameter if you did not want the function to have null access for the initial
POR of the IOCDS.
v Specify at least one LP in the candidate list.
ICP181I UUID STATEMENT SPECIFIED AFTER FUNCTION STATEMENT

Explanation: The UUID statement was specified after a FUNCTION statement was specified. The UUID statement is
optional but if specified must precede all FUNCTION statements.
Programmer response: Move the UUID statement so that it precedes all FUNCTION statements or delete it.
ICP182I INCOMPATIBLE KEYWORDS ON STATEMENT - PCHID AND VCHID

Explanation: You specified incompatible keywords on the CHPID or FUNCTION statement. The PCHID and
VCHID keywords are mutually exclusive.
Programmer response: Specify only one of the keywords for this statement.

ICP183I • ICP188I
ICP183I VCHID KEYWORD NOT ALLOWED ON THIS MACHINE TYPE

Explanation: VCHID keyword supported on M/T 2964 and later, but not on earlier machines.
Programmer response: Correct and re-run IOCP.
ICP184I IF PNETID OF IEDN SPECIFIED, ALL PNETIDS MUST BE IEDN

Explanation: You specified a PNETID value of IEDN for a function type of ROCE. In this case, the following rules
apply:
v All PNETID values specified must be IEDN
v PNETID values prior to the port value that is IEDN cannot be skipped. They must all be specified and must all be
IEDN.
Programmer response: Ensure no port values in the PNETID keyword are skipped and that they are all IEDN.
ICP185I [AID xx | PCHID yyy] PORT z PREVIOUSLY CONNECTED TO CSYSTEM=system-ID-1 CANNOT

ALSO CONNECT TO system-ID-2
Explanation: CSYSTEM is optional on some Coupling CHPIDs, but if specified, first CHPID found must include it,
and all subsequent CHPIDs for a given AID xx or PCHID yyy and port z must match (or not be specified, defaulting
to original).
ICP186I THE CSYSTEM IS OPTIONAL, BUT IF SPECIFIED, IT MUST BE ON FIRST CHPID FOR [AID xx
| PCHID yyy] PORT z
Explanation: CSYSTEM is optional on some Coupling CHPIDs, but if specified, first CHPID found must include it,
and all subsequent CHPIDs for a given AID xx or PCHID yyy and port z must match (or not be specified, defaulting
to original).
ICP187I UID OUT OF ALLOWED RANGE (0001 to FFFF) for UUID PARTITION xxxxxxxx
Explanation: You specified the UID keyword for partition xxxxxxxx which has UUID checking enabled. The UID
keyword must have a 1- to 4-digit hexadecimal value for any partition with UUID checking enabled. UUID checking
is enabled for partitions specified on the UUID statement.
Programmer response: Ensure that the UID keyword value is a maximum of 4 hexadecimal digits or remove the
partition from the UUID statement so that it does not have UUID checking enabled.
ICP188I [FID xxxx | VCHID yyy] PREVIOUSLY DEFINED PNETID zzzzzzzzzzzzzzzz

Explanation: You specified PNETID zzzzzzzzzzzzzzzz but it was previously defined by either FID xxxx or VCHID
yyy. ISM function types and IQD channel path types require that PNETIDs for different VCHIDs be unique.
Programmer response: Ensure that the PNETID values for ISM function types on different VCHIDs are unique and

ICP189I • ICP201I
that the PNETID values for all IQD channel path types are unique.
ICP189I All VFs of the same TYPE=xxxx [PCHID | PCHID PORT | VCHID] must have the same PNETIDs
Explanation: You specified multiple VFs and FIDs for the same function PCHID, PCHID port, or VCHID but they
have different PNETID values. For ISM and ROCE function types, when multiple VFs and FIDs are defined for the
same VCHID or PCHID, the same PNETID values must be specified for all the VFs on that VCHID or PCHID. For
the ROC2 function type, when multiple VFs and FIDs are defined for the same PCHID port, the same PNETID
values must be specified for all VFs on that same PCHID port.
Programmer response: Ensure that all the PNETID values for the VFs and FIDs on the same ISM VCHID, ROCE
PCHID, or ROC2 PCHID port are identical.
ICP190I UID xxxx PREVIOUSLY DEFINED BY FID yyyy FOR UUID PARTITION zzzzzzzz
Explanation: You specified the UID xxxx for partition zzzzzzzz which has UUID checking enabled. However, FID
yyyy also defined UID xxxx for partition zzzzzzzz. Partitions with UUID checking enabled must have unique UID
values. UUID checking is enabled for partitions specified on the UUID statement.
Programmer response: Ensure that the UID keyword values are unique for partitions specified on the UUID
statement or remove the partition from the UUID statement so that it does not have UUID checking enabled.
ICP200I CHPID xx PREVIOUSLY DEFINED IN CSS y

Explanation: The user has defined a channel path (xx) in CSS y to be added to the IOCDS, but an entry for channel
path xx in CSS y exists in the IOCDS.
Programmer response: Ensure that duplicate channel paths in the same CSS are not specified in the CHPID
statements.
ICP201I xxxxxxx KEYWORD NOT ALLOWED ON [CHPID TYPE=yyy | FUNCTION TYPE=yyyy]

Explanation: You specified the xxxxxxx keyword for a TYPE=yyy channel path or TYPE=yyyy function. The
following list indicates valid values:
v The AID keyword is valid only for CIB and CS5 channel paths.
v The CPATH keyword is valid only for ICP, CIB, CL5, and CS5 channel paths.
v The CSYSTEM keyword is valid only for CIB, CL5, and CS5 channel paths.
v The MIXTYPE keyword is valid only for FC and FCP channel paths.
v The PCHID keyword is valid for all channel path types except CIB, CS5, ICP, and IQD.
v The PNETID keyword is valid only for IQD and OSD channel paths and ISM, ROCE, and ROC2 functions.
v The PORT keyword is valid only for CIB, CL5, and CS5 channel paths and HYL and ROC2 functions.
v The VCHID keyword is valid only for IQD channel paths.
v The VF keyword is valid for all function types except RCE.
Programmer response: Either delete keyword xxxxxxx, or change the channel path or function to the correct type.

ICP202I • ICP205I
ICP202I SWITCH KEYWORD NOT ALLOWED ON CHPID TYPE=xxx

Explanation: You specified the SWITCH keyword for a TYPE=xxx channel path. It is valid only for CVC, CBY, CNC,
CTC, FCV, and FC channel paths.
Programmer response: Either delete the SWITCH keyword or change the channel path type.
ICP203I SWITCH xx ALREADY USED BY CHPID yy

Explanation: SWITCH number xx was already defined for channel path yy. FC channel paths (TYPE keyword)
cannot use the same switch number as CVC, CBY, CNC, CTC, and FCV channel paths.
Programmer response: Change the switch number so that a director with FC channel paths has only FC channel
paths.
ICP204I MAXIMUM NUMBER OF CHANNEL PATHS ALLOWED FOR CHPID TYPE HAS BEEN
EXCEEDED [FOR PARTITION xxxxxxxx]
Explanation: The channel path cannot be defined for one of the following reasons:
v The maximum number of CFS, CFR, CFP, CBS, CBR, CBP, CIB, CL5, CS5, and ICP channel paths has been reached
for your CPC.
v The maximum number of CIB, CL5, CS5, and ICP channel paths has been reached for your CPC in partition
xxxxxxxx.
v The maximum number of FC, FCV, and FCP channel paths has been reached for your CPC.
v The maximum number of IQD channel paths has been reached for your CPC.
v The maximum number of FCP, OSC, OSD, OSE, OSM, OSN, and OSX channel paths has been reached for your
CPC.
v The maximum number of OSC, OSD, OSE, OSM, OSN, and OSX channel paths has been reached for your CPC.
v The maximum number of FCP channel paths has been reached for your CPC.
See Table 18 on page 287 to find the maximum number of channel paths that your CPC supports for a particular
channel path type for the entire configuration or for a logical partition.
Programmer response: Adjust the number of channel paths to the number your CPC supports. If you exceeded the
maximum number of channel paths for a partition, use the candidate list of PARTITION or NOTPART keyword on
CHPID statements to adjust the number of channel paths in that partition to the number your CPC supports.
ICP205I TYPE=xxx CHPIDS CANNOT BE SPECIFIED AS SHARED

Explanation: You attempted to define a TYPE=xxx channel path as a shared channel path by specifying the
SHARED, NOTPART, or IOCLUSTER keyword or an access list of more than one logical partition (LP) in the PART
or PARTITION keyword in the CHPID statement. All channel paths can be shared except CBY, CVC, CFR, and CBR.
Programmer response: Either change the channel path type or define an unshared channel path by deleting the
SHARED, NOTPART, or IOCLUSTER keywords and reducing the access list in the PART or PARTITION keyword to
one LP.

ICP206I • ICP210I
ICP206I [TYPE=xxx CHPIDS | TYPE=xxxx FIDS] INVALID FOR THIS PROCESSOR

Explanation: You specified a channel path type or function type for a CPC that does not support it. See Table 19 on
page 291 to determine the channel path types your CPC supports. See Table 20 on page 292 to determine the function
types your CPC supports.
Programmer response: Either change the channel path type or function type to one the CPC supports or change the
SYSTEM keyword value to a CPC that supports the channel path type or function type you want to define.
ICP207I [PCHID xxx | FID yyyy] PREVIOUSLY DEFINED

Explanation: You have defined a physical channel ID (PCHID) xxx or function ID (FID) yyyy to be added to the
IOCDS, but an entry for PCHID xxx or FID yyyy exists in the IOCDS.
Programmer response: Ensure that duplicate PCHIDs or FIDs are not specified in the CHPID or FUNCTION
statements.
ICP208I [PCHID | VCHID] KEYWORD NOT ALLOWED ON CHPID TYPE=xxx

Explanation: You specified the PCHID or VCHID keyword for a TYPE=xxx channel path. The PCHID keyword is
not allowed for ICP, CIB, CS5, and IQD channel paths. The VCHID keyword can only be specified for an IQD
channel path on some CPCs. See Table 23 on page 294 to determine if your CPC requires VCHID for IQD channel
paths.
Programmer response: Either delete the PCHID or VCHID keyword or change the channel path type.
ICP209I TYPE=xxx CHPIDS CANNOT BE DEFINED AS SPANNED

Explanation: You attempted to define a TYPE=xxx channel path as spanned by specifying more than one CSS ID in
the CSS parameter of the PATH keyword in the CHPID statement. All channel paths can be spanned except CNC,
CTC, CVC, CBY, FCV, CFR, and CBR. See “Keyword applicability for channel paths and functions” on page 294 to
determine which channel path types can be spanned on which CPCs.
Programmer response: Either change the channel path type or specify a single CSS ID in the CSS parameter of the
PATH keyword.
ICP210I CHPARM=xx IS INVALID FOR TYPE=yyy CHPIDS

Explanation: You specified the CHPARM keyword with a value of xx for a TYPE=yyy channel path. The CHPARM
keyword is valid only for CNC, FCV, FC, OSD, and IQD channel paths. For CNC, FCV, and FC channel paths, the
valid values are 00 and 01. For OSD channel paths, the valid values are 00 and 02. For IQD channel paths, the valid
values are 00, 40, 80, and C0.
Programmer response: Ensure that CHPARM is specified only for channel path types that support it and specifies
only a valid value for the channel path type.

ICP211I • ICP215I
ICP211I MAXIMUM NUMBER OF CHANNEL PATHS ALLOWED FOR AID xx HAS BEEN EXCEEDED
Explanation: The maximum number of CIB channel paths (TYPE keyword) has been reached for AID xx. You can
specify a maximum of 16 CIB channel paths on an AID.
Programmer response: Reduce the number of channel paths on the AID.
ICP212I **WARNING** CHPARM xx IS IGNORED FOR TYPE=yyy CHPIDS

Explanation: IOCP ignores the CHPARM value for the CHPID type. It substitutes a valid value. See Table 9 on page
77 for additional information.
Programmer response: Correct the CHPARM value.
ICP213I TOO MANY TYPE=xxxx [AIDS | PCHIDS | VCHIDS] DEFINED ('nnnn'X ALLOWED)
Explanation: The channel path or function cannot be defined because too many have already been defined. See the
following:
v The maximum number of CIB or CS5 AIDs has been reached for your CPC. You can define a maximum of nnnn
hexadecimal AIDs for channel path type xxx.
v The maximum number of CL5 PCHIDs has been reached for your CPC. You can define a maximum of nnnn
hexadecimal PCHIDs for channel path type xxx.
v The maximum number of ISM VCHIDs has been reached for your CPC. You can define a maximum of nnnn
hexadecimal VCHIDs for function type ISM.
v The maximum number of RCE, ROCE, or ZEDC PCHIDs has been reached for your CPC. You can define a
maximum of nnnn hexadecimal PCHIDs for function type xxxx.
See Table 18 on page 287 to find the maximum number of channel paths and functions that your CPC supports for
all channel path and function types.
generate an IOCS and does not produce configuration reports.
Programmer response: Adjust the number of AIDs, PCHIDs, or VCHIDs to the number your CPC supports for the
given channel path or function type.
ICP214I CHPARM=xx IS ONLY SUPPORTED ON ONE TYPE=yyy CHPID

Explanation: You specified a CHPARM value of xx for a TYPE=yyy channel path. For IQD channel paths, IOCP
allows only one channel path to have a CHPARM value of x2.
Programmer response: Change the CHPARM keyword value.
ICP215I VF xxx PREVIOUSLY DEFINED BY FID yyyy FOR [PCHID | VCHID] zzz [PORT w]
Explanation: You specified VF xxx but that VF was already defined by FID yyyy on PCHID or VCHID zzz or on
PORT w for PCHID zzz. VF numbers must be unique for a given PCHID or VCHID or for a given port on a PCHID.
Programmer response: Change the VF keyword value.

ICP216I • ICP221I
ICP216I VCHID KEYWORD NOT ALLOWED ON CHPID TYPE=yyy CHPID

Explanation: VCHID keywords are only allowed on a subset of the CHPID types (IQD, ...). See Table 23 on page 294
for more information.
generate an IOCDS and it does not produce configuration reports. IOCP prints this message and then continues
processing.
ICP217I TOO MANY PNETID PARAMETERS SPECIFIED FOR TYPE=xxxx WHICH HAS A MAXIMUM OF
n
Explanation: You specified too many PNETID parameters for channel path or function type xxxx. IOCP limits the
number of PNETID parameters to the number of ports that are available for the given channel path or function type
which is n. However, the ROC2 function supports two ports but each FID is associated with only one of the ports
which is specified with the PORT keyword. Therefore, the ROC2 function type supports only one PNETID for a FID
definition.
Programmer response: Reduce the number of PNETID parameters so they do not exceed the maximum supported.
ICP218I AID xx OUT OF RANGE FOR TYPE=yyy CHPIDS

Explanation: You specified an AID value of xx for a TYPE=yyy channel path. For CIB and CS5 channel paths, the
AID value must be in a range supported by your CPC. See Table 18 on page 287 to determine the AID range
supported by your CPC for the given channel path type.
Programmer response: Change the AID keyword value.
ICP219I FUNCTION STATEMENT FOR PCHID xxx MISSING REQUIRED VF=

Explanation: You specified the FUNCTION statement with a PCHID keyword value of xxx but did not specify the
VF keyword. Some function types (TYPE keyword on FUNCTION statement) require the VF keyword on some CPCs.
See Table 23 on page 294 to determine which function types require the VF keyword on which CPCs.
ICP220I FUNCTION STATEMENT FOR PCHID xxx MISSING REQUIRED TYPE=

Explanation: You specified the FUNCTION statement with a PCHID keyword value of xxx but did not specify the
TYPE keyword. The TYPE keyword on the FUNCTION statement is required on some CPCs. See Table 23 on page
294 to determine which CPCs require the TYPE keyword.
ICP221I VF OUT OF ALLOWED RANGE (1:N) FOR THIS FUNCTION TYPE

Explanation: You specified a VF keyword value that is invalid for the specified function type on this CPC. See
Table 10 on page 106 to determine the valid VF range for this function type on this CPC.

ICP222I • ICP226I
ICP222I FUNCTION TYPE KEYWORD CONFLICT WITH FID xxxx

Explanation: You specified a TYPE keyword value on a FUNCTION statement that conflicts with the TYPE specified
for FID xxxx. All function types for a given PCHID or VCHID must be identical or compatible. See “FUNCTION” on
page 103 TYPE keyword section to determine which function types are compatible.
ICP223I MAXIMUM NUMBER OF CHANNEL PATHS PER PORT, [AID xx | PCHID yyy] HAS BEEN
EXCEEDED
Explanation: You specified too many channel paths on a port of AID xx or PCHID yyy. See Table 17 on page 286 to
determine the maximum number of channel paths for a CS5 AID port or CL5 PCHID port.
ICP224I INVALID PORT VALUE FOR [TYPE=xxx CHPIDS | TYPE=xxxx FIDS]

Explanation: You specified an invalid PORT keyword value for a TYPE=xxx channel path or TYPE=xxxx function.
See Table 17 on page 286 to determine the maximum port number supported by CL5 and CS5 channel paths and
HYL and ROC2 functions.
ICP225I CONTROL UNIT xxxx ALREADY DEFINED ON THE CHANNEL PATH ASSOCIATED WITH
CHPID yy IN CSS z
Explanation: A CNC, CTC, FCV, or FC channel path (TYPE keyword) cannot attach to more than one control unit
unless it attaches to an ESCON or FICON Director (SWITCH keyword) or unless the control units have unique
logical addresses (CUADD keyword). An IQD, OSD, OSM, or OSX channel path (TYPE keyword) cannot have more
than one control unit defined to it unless the control units have unique logical addresses (CUADD keyword). Control
unit xxxx does not have a logical address and channel path yy in CSS z does not attach to a director.
Programmer response: Do one of the following:
v Attach the channel path to an ESCON or FICON Director
v If you are defining a control unit that supports logical addressing, specify the unique logical addresses that your
control unit supports by using the CUADD keyword.
v If you must define more than 256 devices for an IQD channel path or 254 devices for an OSD, OSM, or OSX
channel path, multiple control units must be defined. Specify a unique logical address for each control unit using
the CUADD keyword.
ICP226I LINK ADDRESS MUST BE SUPPLIED. CHPID xx IN CSS y ATTACHES TO SWITCH zz

Explanation: A valid link address (LINK keyword) must be supplied for all channel paths that attach to an ESCON
or FICON Director (SWITCH keyword). A valid link address is a hexadecimal number in the range of 01-FE for CVC,
CBY, CNC, CTC, and FCV channel paths (TYPE keyword). The range is 0000-FFFF for an FC channel path. A link
address of ** or **** is not valid for a channel path that attaches to a director.

ICP227I • ICP231I
Programmer response: Specify a valid link address for all channel paths in the PATH parameter of the CNTLUNIT
statement that attach to a director.
ICP227I CHPID xx IN CSS y HAS NOT BEEN DEFINED

Explanation: The PATH parameter of the CNTLUNIT statement contains an undefined channel path. To eliminate
cascading error messages, IOCP issues this message only if no error has been found while processing CHPID
statements.
Programmer response: If the channel path attached to the control unit is not valid, specify the correct path to the
control unit. If channel path xx in CSS y is valid, define the channel path before you use it in a CNTLUNIT
statement.
ICP228I CUADD SPECIFIED FOR CONTROL UNIT ATTACHING TO TYPE=xxx CHPIDS

Explanation: The CUADD keyword was specified for a control unit attaching to TYPE=xxx channel paths. The
CUADD keyword is valid only for control units that attach to CNC, CTC, FCV, FC, IQD, OSD, OSM, or OSX channel
paths (TYPE keyword).
Programmer response: Either delete the CUADD keyword from the CNTLUNIT statement or change the PATH
keyword values so that the control unit attaches to channel paths which support the CUADD keyword.
ICP229I LINK KEYWORD SPECIFIED FOR CONTROL UNIT ATTACHING TO TYPE=xxx CHPIDS
Explanation: You specified the LINK keyword for a control unit that attaches to TYPE=xxx channel paths. You can
specify the LINK keyword only for control units that attach to CVC, CBY, CNC, CTC, FCV, or FC channel paths
(TYPE keyword).
Programmer response: Either delete the LINK keyword from the CNTLUNIT statement or change the PATH
keyword values so that the control unit attaches to channel paths which support the LINK keyword.
ICP230I CONTROL UNIT xxxx PREVIOUSLY DEFINED

Explanation: The user has defined a control unit (xxxx) to be added to the IOCDS, but an entry for control unit xxxx
exists.
Programmer response: Ensure that duplicate control unit numbers are not specified in the CNTLUNIT statements.
ICP231I MAXIMUM NUMBER OF CONTROL UNITS EXCEEDED

Explanation: You have exceeded the number of PCUs IOCP supports on your CPC. See “Machine limits” on page
286 to check the limits for your CPC.
Programmer response: Reduce the number of physical control units in your I/O configuration.

ICP232I • ICP235I
ICP232I DUPLICATE CHPID NUMBERS SPECIFIED IN CSS x

Explanation: An entry for a control unit to be added contains duplicate CHPIDs in CSS x.
Programmer response: Ensure that channel path identifiers are not duplicated in any CSS in the CNTLUNIT
statement.
ICP233I UNIT ADDRESS SETS EITHER OVERLAP OR GENERATE HEX ADDRESS > FF
Explanation: An entry to add a control unit contains an incorrect unit address. The unit address (with the number
of addresses) either exceeds (hexadecimal) FF or duplicates another unit address for the control unit.
Programmer response: Change the unit addresses specified in the UNITADD parameter so that they are correct.
ICP234I INVALID MIX OF CHPID TYPES WERE SPECIFIED FOR THIS CONTROL UNIT
Explanation: An entry for a control unit contains an incorrect mixture of channel path types (TYPE keyword). All
channel path types attaching to a control unit must meet the following restrictions:
v All CBY or CVC
v CTC
v All CFS or CBS
v FCP
v IQD
v OSC
v OSD
v OSE
v OSM
v OSN
v OSX
Programmer response: Ensure that all of the channel paths attaching to the control unit are allowed to mix.
ICP235I TOO MANY CONTROL UNITS ASSIGNED TO THE CHANNEL PATH ASSOCIATED WITH
CHPID xx IN CSS y
Explanation: The channel path associated with chpid xx in CSS y attaches to too many control units. A parallel
channel path can attach to a maximum of 48 control units. CNC and CTC channel paths can attach to a maximum of
120 control units. IQD channel paths can attach to a maximum of 64 control units. OSD, OSM, and OSX channel
paths can attach to a maximum of 16 control units. FCV and FC channel paths can attach to a maximum of 256
control units. CFS, CFP, CBS, CBP, CIB, CL5, CS5, FCP, ICP, OSC, OSE, and OSN channel paths can attach only to one
control unit.
Programmer response: Correct the I/O configuration so that no channel path attaches to more than the maximum
number of control units allowed for its channel path type.

ICP236I • ICP239I
ICP236I ONLY ONE UNIT ADDRESS RANGE MAY BE SPECIFIED FOR CONTROL UNITS ATTACHING
TO TYPE=xxx CHPIDS
Explanation: The CNTLUNIT statement contains more than one unit address or unit address range on the
UNITADD keyword. Control units attaching to CNC, CTC, FC, or FCV channel paths can have only one range of
unit addresses on the CNTLUNIT statement.
Programmer response: Specify only one unit address range for the control unit.
ICP237I ATTRIBUTES FOR ATTACHED CHPIDS ARE INCONSISTENT

Explanation: You have specified a mixture of shared and unshared channel paths in the PATH keyword of the
CNTLUNIT statement. You cannot attach shared and unshared channel paths to the same control unit. A shared
channel path has either the SHARED or NOTPART keyword specified in the CHPID statement or is spanned or has
an access list of more than one logical partition specified in the PART or PARTITION keyword of the CHPID
statement.
Programmer response: Ensure that the channel paths attaching to the control unit are either all shared or all
unshared.
ICP238I UNIT ADDRESS LIMIT HAS BEEN EXCEEDED FOR THE CHANNEL PATH ASSOCIATED WITH
CHPID xx IN CSS y
Explanation: The channel path associated with chpid xx in CSS y exceeded the maximum number of unit addresses
(devices) allowed for an ESCON or FICON channel path. See Table 17 on page 286 to find the number of unit
addresses (devices) for ESCON or FICON channel paths that your CPC supports.
IOCP might generate more unit addresses for a spanned FC channel path (PATH keyword in the CHPID statement)
than for an unspanned channel path. See “FC control unit definition considerations:” on page 86. CSS y is always the
lowest CSS ID number in the control unit associated with the channel path and is not necessarily the CSS that caused
the error.
Programmer response: Modify your I/O configuration so that the CNC, CTC, FCV, or FC channel path does not
attach to more unit addresses than your CPC supports. You can accomplish this task by doing one or more of the
following:
v Reduce the unit address counts of the control units attached to the channel path
v Reduce the number of control units to which the channel path is attached
v For a spanned FC channel path, use the same link address in each CSS of a control unit for the channel path.
ICP239I TOO MANY CHPIDS ASSIGNED TO CONTROL UNIT

Explanation: A control unit attached to FCP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel paths or an ESCON
CTC or FICON CTC control unit can have only one channel path assigned to it in each CSS. IOCP defines an ESCON
CTC control unit as either of the following:
v A control unit that attaches to only TYPE=CNC or TYPE= FCV channel paths and has UNIT=SCTC coded.
IOCP defines a FICON CTC control unit as a control unit that attaches to only TYPE=FC channel paths and has
UNIT=FCTC coded.
Programmer response: Ensure that the control unit only attaches to one channel path per CSS if the control unit is

ICP240I • ICP244I
an ESCON CTC or FICON CTC control unit or the control unit attaches to FCP, IQD, OSC, OSD, OSE, OSM, OSN, or
OSX channel paths.
ICP240I **WARNING** SHARED KEYWORD IGNORED FOR CONTROL UNITS ATTACHING TO

NON-CVC CHPIDS
Explanation: You specified the SHARED keyword in the CNTLUNIT statement for a control unit that attaches to a
channel path type other than CVC. IOCP processes the SHARED keyword for control units that attach to only CVC
channel paths. IOCP ignores the SHARED keyword for control units attached to any other type of channel path.
System action: IOCP prints this warning message following the CNTLUNIT statement in question and continues
processing.
Programmer response: Either delete the SHARED keyword or change the channel path type.
ICP241I **WARNING** PROTOCL KEYWORD IGNORED FOR CONTROL UNITS ATTACHING TO

NON-CVC CHPIDS
Explanation: You specified the PROTOCL keyword for a control unit that attaches to a channel path type other than
CVC. IOCP processes the PROTOCL keyword for control units that attach to only CVC channel paths. IOCP ignores
the PROTOCL keyword for control units attached to any other type of channel path.
processing.
Programmer response: Either delete the PROTOCL keyword or change the channel path type.
ICP242I LINK ADDRESS xx ALREADY ESTABLISHED FOR CHPID yy IN CSS w BY CONTROL UNIT zzzz
Explanation: A CVC or CBY channel path (TYPE keyword) cannot have more than one link address (LINK
keyword) unless it attaches to an ESCON Director (SWITCH keyword). Channel path yy in CSS w does not attach to
a director and has been assigned link address xx by control unit zzzz. Link address xx differs from the link address
that was specified on the CNTLUNIT statement that is in error.
Programmer response: Remove all link addresses for the channel path, make all link addresses identical, or attach
the channel path to a director.
ICP243I CUADD INCONSISTENCY FOR LINK ADDRESS wwww ON CHPID xx IN CSS m WITH
CONTROL UNIT yyyy DEFINED FOR CHPID zz IN CSS n
Explanation: Two ESCON, FICON, IQD, OSD, OSM, or OSX control units have the same path (css.chpid.link or
switch.link) but only one of the control units has a logical address (CUADD keyword). When a control unit logical
address (CUADD keyword) is defined for a path (css.chpid.link or switch.link), all other occurrences of the path must
also have a control unit logical address.
Programmer response: Either add the CUADD keyword to all control units attaching to the css.chpid.link or
switch.link in question or change the offending control unit so that it does not attach to the same css.chpid.link or
switch.link as control unit yyyy.
ICP244I CUADD ALREADY USED FOR LINK ADDRESS xxxx ON SWITCH yy BY CONTROL UNIT zzzz
Explanation: Two ESCON or FICON control units have the same path (switch.link) through an ESCON or FICON
Director (SWITCH keyword) and the same logical address (CUADD keyword). You cannot attach two ESCON or
FICON control units to the same port on an ESCON or FICON Director and specify the same logical address for each
control unit unless the channel paths are all the same type (TYPE keyword) or a mix of CNC and FCV channel paths
and you are trying to define one of the following:
v ESCON CTC communications
v FICON CTC communications

ICP245I • ICP246I
v A single physical control unit as multiple control units for use by different logical partitions (LPs).
Note: Two control units can never have the same path (css.chpid.link.cuadd) for a given channel path.
v If you are defining two physical control units, correct the switch number or link address for one of the control
units so that the control units attach to different ESCON or FICON Directors or different link addresses on the
same director.
v If you are defining ESCON CTC or FICON CTC communications, ensure that both control units are defined as
ESCON CTC or FICON CTC control units (see “IOCP definitions” on page 56) and that the channel paths are all
the same type or a mix of CNC and FCV channel paths.
v If you are defining a single physical control unit as multiple control units for use by different LPs, ensure that the
control units have the same unit address ranges, the channel paths in each control unit are all the same type or all
CNC and FCV, and none of the paths attaching to the different control units belong to the same LP. The logical
partition intersection test is based on the initial access lists of the channel paths attaching to the control units.
v If you are defining a control unit that supports logical addressing, change the CUADD keyword values to the
unique logical addresses that your control unit supports. Ensure that the channel paths in each control unit are the
same type or a mix of CNC and FCV channel paths.
ICP245I THERE IS NOT A 1-TO-1 CORRESPONDENCE BETWEEN CHPIDS AND LINK ADDRESSES IN
CSS x
Explanation: The LINK parameter for CSS x on the CNTLUNIT statement does not contain the same number of link
addresses as the PATH parameter for CSS x contains channel paths. The number of channel paths must be the same
as the number of link addresses in a CSS.
Programmer response: Ensure that you have entered a valid link address or ** for every channel path in the PATH
parameter for the CSS. Ensure that there are no extraneous channel paths in the PATH parameter and also that there
are no extraneous link addresses in the LINK parameter. Add or remove link addresses or channel paths as necessary.
ICP246I LINK ADDRESS xxxx ALREADY USED FOR SWITCH yy BY CONTROL UNIT zzzz
Explanation: One of the two following conditions exists:
v You have assigned both a parallel and an ESCON control unit to the same switch.link (yy.xx). You cannot attach
parallel and ESCON control units to the same port on an ESCON Director.
v You have assigned two ESCON or FICON control units to the same switch.link (yy.xx or yy.xxxx). You can define
ESCON or FICON control units to the same port on an ESCON or FICON Director only if the channel paths are all
the same type or a mix of CNC and FCV channel paths and you are trying to define one of the following:
– ESCON CTC communications
– FICON CTC communications
– A single physical control unit as multiple control units for use by different logical partitions (LPs)
– A control unit that supports logical addressing (CUADD keyword)
Note: Two control units can never have the same path (css.chpid.link.cuadd) for a given channel path.
Programmer response: For condition 1, do one of the following:
v Attach one of the control units to a different director or link.
v Change the types of channel paths to the control units so that they are all parallel or all CNC, CTC, and FCV.

ICP247I • ICP250I
For condition 2, do one of the following:

v If you are defining two physical control units, correct the switch number or link address for one of the control
units so that the control units attach to different ESCON or FICON Directors or different link addresses on the
same director.
v If you are defining ESCON CTC or FICON CTC communications, ensure that both control units are defined as
ESCON CTC or FICON CTC control units (see “IOCP definitions” on page 56) and that all channel paths are the
same type or a mix of CNC and FCV channel paths.
v If you are defining a single physical control unit as multiple control units for use by different logical partitions
(LPs), ensure that the control units have the same unit address ranges, the channel paths in each control unit are
all the same type or a mix of CNC and FCV channel paths, and none of the paths attaching to the different control
units belongs to the same logical partition. The logical partition intersection test is based on the initial access lists
of the channel paths attaching to the control units.
v If you are defining a control unit that supports logical addressing, specify the unique logical addresses that your
control unit supports by using the CUADD keyword on each of the CNTLUNIT statements, and ensure that the
channel paths in each control unit are all the same type.
ICP247I UNIT ADDRESS xx ALREADY RECOGNIZED BY CONTROL UNIT yyyy FOR LINK ADDRESS zz
ON SWITCH ww
Explanation: Two parallel control units both attach to the same path (switch.link) and have a unit address in
common. Unit address values for the control units must be unique. CVC and CBY control units that attach to the
same switch.link must be daisy-chained. All daisy-chained control units must have unique unit addresses.
Programmer response: Either change the UNITADD keyword values so that they are unique or attach one of the
control units to a different director or link.
ICP248I CUADD ALREADY USED BY CONTROL UNIT xxxx FOR THE CHANNEL PATH ASSOCIATED
WITH CHPID yy IN CSS z
Explanation: Two ESCON, FICON, OSD, or IQD control units both attach to the same channel path associated with
chpid yy in CSS z and the same logical address (CUADD keyword). The CUADD values must be different. This
difference ensures that the combination css.chpid.cuadd is unique for all control units.
Programmer response: Change the CUADD keyword values so that they are unique.
ICP249I **WARNING** UNITADD KEYWORD IGNORED FOR CONTROL UNITS ATTACHING TO

TYPE=xxx CHPIDS
Explanation: You specified the UNITADD keyword for a control unit that attaches to TYPE=xxx channel paths.
IOCP ignores the UNITADD keyword for control units that attach to CFS, CFP, CBS, CBP, ICP, CIB, CL5, CS5, FCP,
IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel paths.
processing.
Programmer response: Either delete the UNITADD keyword or change the channel path type.
ICP250I A CONTROL UNIT CANNOT ATTACH TO TYPE=xxx CHPIDS

Explanation: You specified a control unit that attaches to TYPE=xxx channel paths. You cannot attach a control unit
to CFR or CBR channel paths. CFR or CBR channel paths do not require CNTLUNIT or IODEVICE statements.

ICP251I • ICP255I
Programmer response: Either delete the CNTLUNIT statement or change the channel path type.
ICP251I INVALID CUADD VALUE SPECIFIED FOR CONTROL UNIT ATTACHING TO TYPE=xxx CHPIDS
Explanation: You specified an incorrect control unit logical address (CUADD keyword) for a control unit that
attaches to TYPE=xxx channel paths. CNC, CTC, and FCV channel paths must have a CUADD value in the range
00-0F. FC, IQD, OSD, OSM, and OSX channel paths must have a CUADD value in the range of 00-FF. However, if FC
channel paths are mixed with CNC, or FCV, channel paths, the CUADD value must be in the range 00-0F.
Programmer response: Either correct the CUADD keyword value or change the channel path type.
ICP252I INVALID LINK ADDRESS xxxx SPECIFIED FOR CONTROL UNIT ATTACHING TO TYPE=yyy
CHPIDS
Explanation: You specified an incorrect link address (xxxx) for a control unit that attaches to TYPE=yyy channel
paths. A valid link address range is 01-FE for CBY, CVC, CNC, CTC, and FCV channel paths. The valid link address
range is 0000-FFFF for an FC channel path.
Programmer response: Either correct the LINK keyword value or change the channel path type.
ICP253I LINK ADDRESS MUST NOT BE SPECIFIED. CHPID xx IN CSS y DOES NOT ATTACH TO A
SWITCH
Explanation: You specified a link address other than two or four asterisks (** or ****) for channel path xx in CSS y,
which does not attach to a switch. FC channel paths (TYPE keyword) that do not attach to a switch must either have
no link address or a link address of asterisks.
Programmer response: Delete the LINK keyword or specify a link address of two asterisks for all FC channel paths
that do not attach to a switch.
ICP254I UNIT=FCTC INVALID FOR CONTROL UNITS ATTACHING TO TYPE=xxx CHPIDS

Explanation: You specified control unit type FCTC for a control unit attached to TYPE=xxx channel paths. FCTC is
used to specify a FICON CTC control unit. FICON CTC control units require FC channel paths.
System action: IOCP prints this message following the CNTLUNIT statement in question and continues processing.
Programmer response: Change the UNIT keyword value or attach the control unit to only FC channel paths.
ICP255I LINK xxxx IS INCONSISTENT WITH A ONE-BYTE | TWO-BYTE LINK ALREADY SPECIFIED
FOR THE CHANNEL PATH ASSOCIATED WITH CHPID yy IN CSS z
Explanation: You specified link address xxxx, which is inconsistent with a link address already specified for the
channel path associated with chpid yy in CSS z. Link addresses for an FC channel path (TYPE keyword) must be all
1-byte (2-digit) or 2-byte (4-digit) values.
Programmer response: Change the link address values for the FC channel path so that they are all the same size.
See “Defining FICON switch configurations” on page 38 for more information about specifying link addresses for FC
channel paths.

ICP256I • ICP300I
ICP256I INCOMPATIBLE CONTROL UNIT TYPES FOR THE CHANNEL PATH ASSOCIATED WITH
CHPID xx IN CSS y
Explanation: The two control units on the channel path associated with chpid xx in CSS y are incompatible. One is a
FICON CTC control unit and the other is not. For FC channel paths that do not attach to a switch, all the control
units on the channel path must either be FICON CTC control units or none of them can be FICON CTC control units.
IOCP defines a FICON CTC control unit as a control unit that attaches to only TYPE=FC channel paths and has
UNIT=FCTC coded.
Programmer response: Ensure all the control units on the FC channel path are FCTC or none of them are FCTC.
ICP257I ALL TYPE=CIB|CL5|CS5 CHPIDS ATTACHED TO CONTROL UNIT MUST CONNECT TO THE
SAME SYSTEM
Explanation: You defined a control unit with CIB, CL5, or CS5 channel paths (TYPE keyword) where all of the
channel paths do not connect to the same target CPC (CSYSTEM keyword on CHPID statement). All CIB, CL5, and
CS5 channel paths in a control unit must have identical CSYSTEM keyword values.
Programmer response: Ensure that all of the CIB, CL5, and CS5 channel paths attaching to the control unit all have
the same CSYSTEM keyword value.
ICP258I TYPE=xxx CHPIDS NOT CONNECTED TO THE LOCAL SYSTEM CANNOT BE MIXED WITH
TYPE=ICP CHPIDS
Explanation: You defined a control unit with CIB, CL5, CS5, and ICP channel paths (TYPE keyword). ICP channel
paths can be used only for internal coupling connections. However, the CIB, CL5, or CS5 channel paths do not
connect to the local system because their CSYSTEM keyword values (CHPID statement) do not match the local
system (LSYSTEM keyword on ID statement) for this I/O configuration.
Programmer response: Ensure that all of the CIB or CS5 channel path attaching to the control unit have CSYSTEM
keyword values that specify the local system (LSYSTEM keyword) or remove all ICP channel paths from the control
unit.
ICP259I UNIT=STP INVALID FOR CONTROL UNITS ATTACHED TO INTERNALLY COUPLED CHPIDS
Explanation: You defined a control unit for Server Time Protocol (STP) to provide a timing network between CPCs.
ICP and internally coupled CIB, CL5, and CS5 channel paths (TYPE keyword) cannot be used for STP. CIB, CL5, and
CS5 channel paths have an internal coupling connection when their CSYSTEM keyword values (CHPID statement)
match the local system name (LSYSTEM keyword on ID statement). See “Server Time Protocol” on page 30 for more
information about STP.
Programmer response: Remove all ICP channel paths from the control unit. Also, remove the CIB, CL5, or CS5
channel paths from the control unit or connect the CIB, CL5, and CS5 channel paths to the external CPC by changing
all of their CSYSTEM keyword values.
ICP300I UNIT ADDRESS EXCEEDS HEX ADDRESS FF

Explanation: The keyword value for the unit address (specified in either the UNITADD or the ADDRESS keyword
in the IODEVICE statement) plus the keyword value for count (specified in the ADDRESS keyword), minus one,
exceeds the maximum allowable value of (hexadecimal) FF for a unit address.

ICP301I • ICP305I
Programmer response: Correct the keyword values.
ICP301I DEVICE xxxx HAS EXCEEDED THE MAXIMUM NUMBER OF I/O DEVICES ALLOWED FOR
SUBCHANNEL SET z IN CSS y
Explanation: Device number xxxx has exceeded the maximum number of I/O devices supported in subchannel set z
in CSS y. For a dynamic-capable IOCDS, IOCP enforces the MAXDEV keyword value specified for subchannel set z
in CSS y. If no value was specified or if the IOCDS is not a dynamic-capable, IOCP enforces the subchannel set
device limit that the CPC supports. See “Machine limits” on page 286 to check the device limits the CPC supports.
Programmer response: Do one or both of the following:
v Reduce the number of I/O devices in the configuration for the subchannel set for the CSS
v For a dynamic-capable IOCDS, increase the value specified with the MAXDEV keyword for the subchannel set for
the CSS up to the device limit the CPC supports.
ICP302I DUPLICATE CONTROL UNIT NUMBERS SPECIFIED

Explanation: An I/O device contains duplicate control unit numbers.
Programmer response: Ensure that control unit numbers are not duplicated in the IODEVICE statement.
ICP303I UNIT ADDRESS yy DOES NOT EXIST IN CONTROL UNIT xxxx

Explanation: An I/O device contains unit address yy, which does not exist in the range of unit addresses for control
unit xxxx.
Programmer response: Ensure that the correct range of unit addresses is specified in the CNTLUNIT statement and
the IODEVICE statement. Control units attaching to some channel path types (for example OSD channel paths) have
a fixed unit address range that requires you to correct the range of unit addresses specified in the IODEVICE
statement.
ICP304I TOO MANY CHPIDS ASSIGNED TO I/O DEVICE IN CSS x

Explanation: You have assigned too many channel paths in CSS x to the I/O device based on the channel paths in
CSS x in the control units that attach to the device. You can assign a maximum of eight channel paths to a device in a
CSS.
Programmer response: Modify the I/O configuration so that the device does not attach to too many paths in any
CSS.
ICP305I ATTRIBUTES FOR ATTACHED CONTROL UNITS ARE INCONSISTENT

Explanation: An I/O device has specified two or more control units and the control units do not have the same
protocol or are not the same type (1 or 2).
Programmer response: Ensure that the control units are correctly specified in the CNTLUNIT statements and that
the I/O device is correctly specified in the IODEVICE statement.

ICP306I • ICP310I
ICP306I I/O DEVICE CONNECTED TO MORE THAN ONE CONTROL UNIT ON CHPID xx IN CSS y
Explanation: An I/O device has specified two or more control units, and two or more of the control units are
attached to the same channel path xx in CSS y. When a device is assigned to more than one control unit, each control
unit must be attached to a different channel path in a CSS.
Programmer response: Ensure that the control units are correctly specified in the CNTLUNIT statements and that
the I/O device is correctly specified in the IODEVICE statement.
ICP307I NO PATH TO CHPID xx IN CSS y FOR THIS DEVICE

Explanation: IOCP did not find a path from the device to the channel path that was specified by the PATH value in
the IODEVICE statement.
Programmer response: Ensure that the value specified in the PATH parameter is correct. Check the associated
CNTLUNIT and CHPID statements to ensure that there is a path from the device through a control unit to the
specified channel path and CSS.
ICP308I CONTROL UNIT NUMBER xxxx HAS NOT BEEN DEFINED

Explanation: Control unit xxxx has not been defined by a CNTLUNIT statement. To eliminate cascading error
messages, IOCP issues this message only if no error has been found while processing CHPID and CNTLUNIT
statements.
Programmer response: Define control unit xxxx before the control unit is included in an IODEVICE statement.
ICP309I NONE OF THE PARTITIONS IN THE DEVICE CANDIDATE LIST IN CSS x CAN BE ACCESSED
BY THE CHPIDS ASSIGNED TO THIS DEVICE
Explanation: You specified the PART, PARTITION, or NOTPART keyword in the IODEVICE statement for this
device but none of the logical partitions (LPs) in the device candidate list in CSS x can be accessed by the channel
paths in CSS x assigned to this device. This situation indicates that there is no match between the LPs in the
combined candidate lists of the channel paths assigned to the device and the LPs in the device candidate list. If there
is no match, there can be no channel path access to the device in the CSS.
Programmer response: Ensure that the PART, PARTITION, or NOTPART keyword in the IODEVICE statement
generates a device candidate list with at least one LP that is also in the candidate list of one of the channel paths
assigned to the device for the CSS.
ICP310I **WARNING** TIMEOUT KEYWORD IGNORED FOR DEVICES ASSIGNED TO NON-CVC

CHPIDS
Explanation: You specified the TIMEOUT keyword in the IODEVICE statement for a device assigned to channel
paths that are not CVC (TYPE keyword). IOCP processes the TIMEOUT keyword for devices assigned to only CVC
channel paths. IOCP ignores the TIMEOUT keyword for devices assigned to any other type of channel path.
System action: IOCP prints this warning message following the IODEVICE statement in question and continues
processing.
Programmer response: Either delete the TIMEOUT keyword or change the channel path type.

ICP311I • ICP313I
ICP311I CONTROL UNITS FOR THIS DEVICE CONTAIN AN INVALID MIX OF CHPID TYPES
Explanation: An entry for an I/O device contains an incorrect mix of channel path types (TYPE keyword) that are
associated with two or more control units. All channel path types assigned to a device must meet the following
restrictions:
v All CBY or CVC
v All CNC or FCV or FC
v CTC
v All CFS or CBS
v All CFP, CBP, ICP, CL5, CS5, or CIB
v FCP
v IQD
v OSC
v OSD
v OSE
v OSM
v OSN
v OSX
Programmer response: Ensure that all channel paths attaching to control units for the I/O device are allowed to
mix.
ICP312I TOO MANY CONTROL UNITS ASSIGNED TO DEVICE

Explanation: The device is attached to more than one control unit and one of the control units is either an ESCON
CTC or FICON CTC control unit or a control unit attached to CFS, CFP, CBS, CBP, CIB, CL5, CS5, FCP, ICP, IQD,
OSC, OSD, OSE, OSM, OSN, or OSX channel paths. A device can only attach to one control unit when the control
unit is an ESCON CTC or FICON CTC control unit or when the control unit attaches to CFS, CFP, CBS, CBP, CIB,
CL5, CS5, FCP, ICP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel paths.
An ESCON CTC control unit is defined as one of the following:
v A control unit that attaches to only TYPE=CNC or TYPE= FCV channel paths and has UNIT=SCTC coded.
A FICON CTC control unit is defined as a control unit that attaches to only TYPE=FC channel paths and has
UNIT=FCTC coded.
Programmer response: If the device is meant to attach to an ESCON CTC or FICON CTC control unit or to a
control unit attached to CFS, CFP, CBS, CBP, CIB, CL5, CS5, FCP, ICP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX
channel paths, delete all other control units from the IODEVICE statement. Otherwise, delete the ESCON CTC or
FICON CTC control unit or the control unit attached to the CFS, CFP, CBS, CBP, CIB, CL5, CS5, FCP, ICP, IQD, OSC,
OSD, OSE, OSM, OSN, or OSX channel paths from the IODEVICE statement.
ICP313I ATTRIBUTES FOR CHPIDS ASSIGNED TO THIS DEVICE ARE INCONSISTENT

Explanation: You have specified control units in the CUNUMBR keyword of the IODEVICE statement that attach to
a mixture of shared and unshared channel paths. You cannot assign shared and unshared channel paths to the same
device. A shared channel path has either the SHARED or NOTPART keyword specified in the CHPID statement or is
spanned or has an access list of more than one logical partition specified in the PART or PARTITION keyword of the
CHPID statement.

ICP314I • ICP318I
Programmer response: Ensure that the channel paths attaching to the control units for the device are either all
shared or all unshared.
ICP314I **CAUTION** DEVICE CANNOT BE ACCESSED BY PARTITION xxxxxxxx IN CSS y

Explanation: You specified the PART, PARTITION, or NOTPART keyword in the IODEVICE statement and logical
partition xxxxxxxx is in the device candidate list for CSS y. However, the logical partition (LP) cannot access the
device because none of the channel paths in CSS y assigned to the device have access to the LP. That is, none of the
channel paths assigned to the device have the LP in their chpid candidate lists for CSS y.
System action: IOCP prints this caution message following the IODEVICE statement in question and continues
processing.
Programmer response: Ensure that the LP name belongs in the device candidate list and that the channel paths
assigned to the device are not to have access to the LP.
ICP315I NOTPART AND PARTITION KEYWORDS ARE NOT ALLOWED FOR DEVICES ASSIGNED TO
UNSHARED CHPIDS
Explanation: You specified the PART, PARTITION, or NOTPART keyword in the IODEVICE statement of a device
that is assigned to unshared channel paths. You can specify only the PART, PARTITION, or NOTPART keyword in
the IODEVICE statement for devices assigned to shared channel paths. A shared channel path has either the
SHARED or NOTPART keyword specified in the CHPID statement or is spanned or has an access list of more than
one logical partition in the PART or PARTITION keyword of the CHPID statement.
Programmer response: Either delete the PART, PARTITION, or NOTPART keyword in the IODEVICE statement or
assign the device to shared channel paths. You cannot mix shared and unshared channel paths to the same device.
ICP316I **WARNING** PATH KEYWORD IGNORED FOR DEVICES ASSIGNED TO TYPE=xxx CHPIDS
Explanation: You specified the PATH keyword in the IODEVICE statement for a device assigned to TYPE=xxx
channel paths. The PATH keyword is meaningful only for devices assigned to CBY, CVC, CNC, FCV, or FC channel
paths. IOCP ignores the PATH keyword for all other channel path types.
processing.
Programmer response: Either delete the PATH keyword in the IODEVICE statement or change the channel path
type.
ICP317I **WARNING** STADET KEYWORD IGNORED FOR DEVICES ASSIGNED TO TYPE=xxx CHPIDS
Explanation: You specified the STADET keyword in the IODEVICE statement for a device assigned to TYPE=xxx
channel paths. The STADET keyword is meaningful only for devices assigned to CBY, CVC, CNC, CTC, FCV, or FC
channel paths. IOCP ignores the STADET keyword for all other channel path types.
processing.
Programmer response: Either delete the STADET keyword in the IODEVICE statement or change the channel path
type.
ICP318I NOTPART AND PARTITION KEYWORDS ARE NOT ALLOWED FOR DEVICES ASSIGNED TO
TYPE=xxx CHPIDS
Explanation: You specified the NOTPART, PART, or PARTITION keyword in the IODEVICE statement for a device
assigned to TYPE=xxx channel paths. The NOTPART and PART or PARTITION keywords in the IODEVICE
statement are allowed only for CNC, CTC, FCV, FC, FCP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel paths.

ICP319I • ICP322I
Programmer response: Either delete the NOTPART or PART or PARTITION keyword in the IODEVICE statement or
change the channel path type.
ICP319I **WARNING** UNITADD KEYWORD IGNORED FOR DEVICES ASSIGNED TO TYPE=xxx

CHPIDS
Explanation: You specified the UNITADD keyword in the IODEVICE statement for a device assigned to TYPE=xxx
channel paths. IOCP ignores the UNITADD keyword for devices assigned to CFS, CFP, CBS, CBP, ICP, CIB, CL5, CS5,
channel paths.
processing.
Programmer response: Either delete the UNITADD keyword in the IODEVICE statement or change the channel
path type.
ICP320I **WARNING** UNIT=xxxx EXPECTED FOR DEVICES WITH UNIT ADDRESS yy ASSIGNED TO
TYPE=zzz CHPIDS
Explanation: You specified a device type other than xxxx (UNIT keyword) for a device with unit address yy and
assigned to TYPE=zzz channel paths. For OSD, OSE, OSM, and OSX channel paths, IOCP expects devices with unit
addresses in the range 00-FD to have a value of OSA and the device with unit address FE to have a value of OSAD.
For OSN channel paths, IOCP expects the device with unit address FE to have a value of OSAD.
processing.
Programmer response: Change the UNIT keyword value.
ICP321I DEVICE CANNOT BE ACCESSED BY CSS x

Explanation: You specified CSS x in the PATH, PART, PARTITION, or NOTPART keyword in the IODEVICE
statement. However, none of the control units attached to the device have channel paths in that CSS.
Programmer response: Ensure that the CSS ID specified is correct. Check the associated CNTLUNIT and CHPID
statements to ensure that there is a channel path from the CSS through a control unit to the device.
ICP322I NULL DEVICE CANDIDATE LIST SPECIFIED FOR EACH CSS

Explanation: You specified a null device candidate list for every CSS to which this device has access. A device must
be accessible by at least one CSS.
Programmer response: Either delete the device or modify the device candidate list so that at least one CSS can
access the device.
For ICP IOCP 1.1.0:

ICP322I NULL DEVICE CANDIDATE LIST SPECIFIED IN CSS x
Explanation: You cannot specify a null device candidate list for a device. You generated the null device candidate
list by specifying all logical partitions (LPs) in the configuration in the device candidate list parameter of the
NOTPART keyword in CSS x for the device. A device requires at least one LP in its generated device candidate list
for each CSS for which it has channel paths.
Programmer response: Delete one or more of the LP names in the CSS from the device candidate list parameter of
the NOTPART keyword in the IODEVICE statement.

ICP323I • ICP327I
ICP323I **CAUTION** DEVICE WITH UNIT ADDRESS FE REQUIRES A CONTROL UNIT WITH
CUADD=0 TO BE USABLE
Explanation: You specified a device with unit address FE on a control unit with a nonzero CUADD keyword value.
For an OSD, OSM, or OSX channel path, the device with unit address FE (usually specified with UNIT=OSAD on the
IODEVICE statement) can be used only when specified for a control unit without the CUADD keyword or with a
CUADD keyword value of zero.
System action: IOCP prints this caution message following the IODEVICE statement in question and continues
processing.
Programmer response: Ensure that devices with unit address FE on OSD, OSM, or OSX channel paths are assigned
only to control units without the CUADD keyword or with a CUADD keyword value of zero.
ICP324I DEVICE CANDIDATE LIST PREVENTS CONTROL UNIT xxxx FROM ACCESSING DEVICE
Explanation: You specified a null device candidate list for every CSS with access to control unit xxxx. Therefore, the
control unit cannot access the device. Each control unit assigned to the device must be able to access the device.
Programmer response: Either remove the control unit from the CUNUMBR keyword for the device or modify the
device candidate list so that the control unit can access the device from at least one CSS.
ICP325I DEVICES ASSIGNED TO TYPE=xxx CHPIDS CANNOT BE DEFINED IN SUBCHANNEL SET y

Explanation: You specified subchannel set y in the SCHSET keyword in the IODEVICE statement for a device
assigned to TYPE=xxx channel paths. You can specify only subchannel set 0 for devices assigned to CBY, CVC, CFS,
CFP, CBS, CBP, ICP, CIB, CL5, and CS5.
v Delete the SCHSET keyword in the IODEVICE statement.
v Specify only subchannel set ID 0.
v Change the channel path type.
ICP326I UNIT=xxxxxxxx IS NOT ALLOWED FOR DEVICES [WITH UNIT ADDRESS yy] ASSIGNED TO
TYPE=zzz CHPIDS
Explanation: You specified device type xxxxxxxx (UNIT keyword) for a device assigned to TYPE=zzz channel paths.
For OSN channel paths, IOCP requires devices with unit addresses in the range 00-FD to have a value of OSN or
3745.
ICP327I **WARNING** NULL DEVICE CANDIDATE LIST SPECIFIED FOR CSS x SO THE CSS IS
IGNORED BY OTHER KEYWORDS
Explanation: You specified CSS x in the PATH or SCHSET keyword in the IODEVICE statement. However, you
specified a null device candidate list with the PART, PARTITION, or NOTPART keyword in the IODEVICE statement
for that same CSS, which prevents the device from being defined in that CSS. Therefore, IOCP ignores any CSS x
parameters for the PATH and SCHSET keywords for the IODEVICE statement.
processing.
Programmer response: Either correct the CSS parameter in the PATH or SCHSET keyword, or do not specify a null
device candidate list for the CSS.

ICP328I • ICP353I
ICP328I **WARNING** UNIT=xxx IS EXPECTED FOR DEVICES ASSIGNED TO TYPE=zzz CHPIDS

Explanation: You specified a device type xxx (UNIT keyword) for a device assigned to TYPE=zzz channel paths. For
IQD channel paths, IOCP expects the device type to have a value of IQD.
processing.
ICP350I MINIMUM CONFIGURATION WAS NOT DEFINED

Explanation: To generate an IOCDS, ICP IOCP 1.1.0 requires the following:
v For a dynamic-capable configuration (a configuration built by HCD or by specifying IOCP CMS utility option DYN
for VM), the following statements must be true:
– At least one logical partition must have a device defined on a CVC, CBY, CNC, FCV, FC, or FCP channel path
(TYPE keyword) or
– If no device is defined on one of these channel path types, all logical channel subsystems (CSSs) must have an
external coupling link capable of receiving messages defined (CFR, CFP, CBR, or CBP channel path)
– For a configuration that is not dynamic-capable, one of the following statements must be true for each defined
CSS:
- A device is defined on a CVC, CBY, CNC, FCV, FC, or FCP channel path (TYPE keyword)
- An external coupling link capable of receiving messages is defined (CFR, CFP, CBR, or CBP channel path)
- An internal coupling link is defined (ICP channel path). But at least one CSS must have a device defined on a
CVC, CBY, CNC, FCV, FC, or FCP channel path (TYPE keyword).
System action: IOCP does not complete the generation of an IOCDS and does not produce configuration reports.
Programmer response: Ensure that a minimum configuration is defined.
ICP351I ATTEMPTED TO ASSIGN PHYSICAL CONTROL UNIT xxxx TO TWO CONTROL UNIT
HEADERS
Explanation: IOCP attempted to assign physical control unit xxxx to two control unit headers. A logical error has
occurred in IOCP.
Programmer response: Analyze the control unit header that contains the physical control unit. If the control unit
header is valid, save the console sheet and all associated output and notify your service representative.
ICP352I DEVICE NUMBER wwww IN SUBCHANNEL SET n DUPLICATES UNIT ADDRESS xx IN CSS z
ON CONTROL UNIT yyyy
Explanation: When processing device number wwww in subchannel set n, IOCP detected unit address xx duplicated
on control unit yyyy in CSS z. This error occurs if two or more devices with the same unit address are assigned to the
same control unit in the same CSS.
Programmer response: Ensure that unit addresses are not duplicated in the same CSS in IODEVICE statements that
attach to control unit yyyy.
ICP353I IOCP HAS GENERATED xxxxx CONTROL UNIT HEADERS FOR CSS y WHICH EXCEEDS THE
MAXIMUM NUMBER ALLOWED
Explanation: IOCP tried to generate more control unit headers (CUHs) in a CSS than your CPC supports. Review
the limits and rules for the generation of CUHs for your CPC. See “Machine limits” on page 286 to verify what your
CPC supports.
Programmer response: Ensure that the configuration does not exceed the maximum number of CUHs for a CSS.

ICP354I • ICP355I
You might need to reduce the number of physical control units or have more control units share devices.
ICP354I **CAUTION** MINIMUM NUMBER OF DEVICES HAVE NOT BEEN DEFINED FOR [CHPID xx
IN CSS y | THE CHANNEL PATHS IN CSS x IN CONTROL UNIT yyyy]
Explanation: You did not define the minimum number of devices recommended for a coupling facility channel path
or control unit. CFP, CBP, and ICP channel paths (TYPE keyword) assigned to a control unit must have a minimum
of seven devices each. CIB channel paths assigned to a control unit must have a minimum of seven devices each
which is what IOCP tests but some adapters have a minimum of 32 devices. CL5 channel paths assigned to a control
unit must have a minimum of eight devices each, but IBM recommends that 32 be defined. CS5 channel paths
assigned to a control unit must have a minimum of eight devices each. CFS and CBS channel paths must have a
minimum of two devices each. If the message is for channel path xx in CSS y, the channel path is a CFS or CBS
channel path and does not have a control unit. If the message is for the channel paths in CSS x in control unit yyyy,
the minimum number of devices necessary to fully support the channel paths in CSS x in the control unit were not
defined. See “Coupling facility channels” on page 27 section to determine the correct number of devices to define for
your hardware type.
System action: IOCP prints this caution message and continues processing.
Programmer response: Define the minimum number of devices recommended for the channel path type for either
the channel path or the CSS in the control unit.
ICP355I CHANNEL PATH ASSOCIATED WITH CHPID xx IN CSS y HAS zzzzz VALID SUBCHANNELS
WHICH EXCEEDS MAXIMUM ALLOWED [FOR {OSN | 3745} DEVICES]
Explanation: You specified too many valid subchannels for channel path xx in CSS y. Valid subchannels are devices
which are defined to a logical partition and permitted for use by the logical partition. Valid subchannels are
calculated as follows:
v For unshared channel paths and for reconfigurable FCP channel paths on CPCs without fixed HSA, each device
defined in IOCP represents a single valid subchannel. See Table 16 on page 285 to determine if your CPC has fixed
HSA.
v For reconfigurable FCP channel paths on CPCs with fixed HSA, each FCP device defined in IOCP represents
multiple valid subchannels based on the total number of logical partitions (LPs) in its CHPID candidate list.
v For shared channel paths, each device defined in IOCP represents multiple valid subchannels based on the total
number of LPs in its device candidate list.
One of the following conditions exists:

v You defined too many OSA valid subchannels on an OSD, OSM, or OSX channel path. A maximum of 480 OSA
valid subchannels can be defined on an OSD or OSX channel path when priority specification is enabled
(CHPARM=00 is specified on CHPID statement). A maximum of 1920 OSA valid subchannels can be defined on an
OSD, OSM, or OSX channel path when priority specification is disabled (CHPARM=02 is specified). An OSA
subchannel is a device with a unit address in the range 00-FD. The device with unit address FE (OSAD) is
excluded from the maximum.
v You defined too many OSN or 3745 valid subchannels on an OSN channel path. A maximum of 480 OSN valid
subchannels and 180 3745 valid subchannels can be defined on an OSN channel path. OSN subchannels are
defined with UNIT=OSN on the IODEVICE statement, and 3745 subchannels are defined with UNIT=3745. Valid
subchannels are devices with unit addresses in the range 00-FD. The OSAD device with unit address FE is
excluded from the maximums
v You defined too many valid subchannels on an FCP channel path. A maximum of 480 valid subchannels can be
defined on an FCP channel path.
Programmer response: For an unshared channel path or reconfigurable FCP channel paths on CPCs without fixed
HSA, reduce the number of devices defined on the FCP channel path or defined on the OSD, OSM, OSN, or OSX
channel path with unit addresses in the range 00-FD.
For a reconfigurable FCP channel path on CPCs with fixed HSA, do one or both of the following:
v Reduce the number of devices defined on the FCP channel path
v Specify a CHPID candidate list for the FCP channel path with fewer logical partitions.

ICP356I • ICP359I
For a shared channel path, do one or both of the following:

v Reduce the number of devices defined on the FCP channel path or defined on the OSD, OSM, OSN, or OSX
channel path with unit addresses in the range 00-FD.
v Specify device candidate lists with fewer logical partitions for the devices defined on the FCP channel or defined
on the OSD, OSM, OSN, or OSX channel path with unit addresses in the range 00-FD. For devices assigned to
OSN channel paths, specify device candidate lists that restrict the OSN devices to CCL images and the 3745
devices to operating system images.
For an OSD channel path on an OSA-Express2 or later feature or for an OSM or OSX channel path on an
OSA-Express3 or later feature, if you want to disable priority specification to increase the maximum number of valid
subchannels available on the channel, specify the appropriate CHPARM keyword value for the OSD or OSX channel
path (CHPARM=02).
ICP356I IOCP HAS GENERATED xxxxxx VALID SUBCHANNELS FOR TYPE=IQD CHPIDS WHICH
EXCEEDS THE MAXIMUM NUMBER ALLOWED
Explanation: You specified too many valid subchannels for all TYPE=IQD channel paths in your configuration. A
maximum of 12288 valid subchannels can be defined on all IQD channel paths. Valid subchannels are devices that
are defined to a logical partition and permitted for use by the logical partition. Valid subchannels are calculated as
follows:
v For shared channel paths, each device defined in IOCP represents multiple valid subchannels based on the total
number of logical partitions (LPs) in its device candidate list.
Programmer response: For unshared channel paths, reduce the number of devices defined on IQD channel paths.
For shared IQD channel paths, do one or both of the following:
v Reduce the number of devices defined.
v Specify device candidate lists with fewer logical partitions for the devices defined.
ICP357I TOO MANY LINK ADDRESSES ASSIGNED TO THE CHANNEL PATH ASSOCIATED WITH
CHPID xx IN CSS y
Explanation: You have assigned too many link addresses to channel path xx in CSS y. CPCs limit the number of link
addresses you can assign to CNC, CTC, FCV, and FC channel paths. See “Machine limits” on page 286 to verify what
your CPC supports.
Programmer response: Reduce the number of link addresses to which channel path xx is assigned.
ICP358I CONTROL UNIT xxxx DUPLICATES UNIT ADDRESS yy ON CHPID zz IN CSS w

Explanation: When processing control unit xxxx, IOCP detected unit address yy duplicated on parallel channel path
zz in CSS w. This error occurs if two or more control units with the same unit address attach to the same parallel
channel path.
Programmer response: Ensure that control unit xxxx does not duplicate unit addresses for parallel channel path zz.
ICP359I TOO MANY UNIT ADDRESS RANGES ASSIGNED TO THE CHANNEL PATH ASSOCIATED
WITH CHPID xx IN CSS y
Explanation: One of the following conditions exist:
v Channel path xx in CSS y is a CBY or CVC channel path and has exceeded the maximum number of unit address
ranges allowed for the channel path. The maximum number of unit address ranges for CNTLUNIT statements that
attach to a CBY or CVC channel path is 64.

ICP360I • ICP361I
v The maximum number of unit address ranges for control units that attach to shared CTC channel paths (SHARED
keyword on CHPID statement) is 120. IOCP might generate more unit address ranges for control units attaching to
shared CTC channel paths than for other channel path types or unshared CTC channel paths. See “Shared ESCON
CTC considerations” on page 25 for more information.
v Channel path xx in CSS y is a spanned FC channel path (PATH keyword in the CHPID statement) and has
exceeded the maximum number of unit address ranges allowed for the channel path. The maximum number of
unit address ranges for CNTLUNIT statements that attach to an FC channel path is 256. IOCP might generate more
unit address ranges for a spanned FC channel path than for an unspanned channel path. See “FC control unit
definition considerations:” on page 86. CSS y is always the lowest CSS ID number for the channel path and is not
the only CSS that caused the error.
Programmer response: For a CBY or CVC channel path, reduce or consolidate the unit address ranges for
CNTLUNIT statements attached to channel path xx in CSS y. For a shared CTC channel path, do one or more of the
following:
v Reduce the number of control units attached to channel path xx
v For the devices assigned to channel path xx, reduce the number of logical partitions (LPs) in their device candidate
lists (PART, PARTITION, or NOTPART keyword on IODEVICE statement). The default device candidate list for a
device assigned to a shared CTC channel path contains all the LPs in the candidate list of the CTC channel path.
Specify a subset of these LPs in the PART, PARTITION, or NOTPART keyword in the IODEVICE statements for
each device attaching to the same control unit.
For an FC channel path, do one or more of the following:

v Reduce the number of control units attached to channel path xx
v Use the same link address in each CSS of a control unit for channel path xx.
ICP360I UNIT ADDRESS LIMIT FOR THE CHANNEL PATH ASSOCIATED WITH CHPID xx IN CSS y
HAS BEEN EXCEEDED
Explanation: The maximum number of unit addresses (UNITADD keyword on CNTLUNIT statement) for the
control units that attach to FC channel path xx or shared CTC path xx has been exceeded. See Appendix F, “Machine
limits and rules,” on page 285 to check the device (unit address) limit allowed for FC or CTC channel paths for your
CPC.
IOCP might generate more unit addresses for physical control units attaching to shared CTC channel paths than for
other channel path types or unshared channel paths. See “Shared ESCON CTC considerations” on page 25 for more
information.
IOCP might generate more unit addresses for a spanned FC channel path (PATH keyword in the CHPID statement)
than for an unspanned channel path. See “FC control unit definition considerations:” on page 86. CSS y is always the
lowest CSS ID number for the channel path and is not the only CSS that caused the error.
Programmer response: Do one or more of the following:
v Reduce the unit address counts of the control units attached to channel path xxin CSS y
v Reduce the number of control units attached to channel path xx in CSS y
v For a shared CTC channel path, reduce the number of logical partitions in their device candidate lists (PART,
PARTITION, or NOTPART keyword on the IODEVICE statement) for the devices assigned to the channel path. The
default device candidate list for a device assigned to a shared CTC channel path contains all the LPs in the
candidate list of the CTC channel path. Specify a subset of these partitions in the PART, PARTITION, or NOTPART
keyword on the IODEVICE statements for each device attaching to the same control unit.
v For a spanned FC channel path, use the same link address in each CSS of a control unit for channel path xx.
ICP361I DEVICE NUMBER xxxx IN SUBCHANNEL SET w ASSIGNED TO CONTROL UNIT yyyy IS
DUPLICATED IN A LOGICAL PARTITION IN CSS z
Explanation: Device number xxxx in subchannel set w assigned to control unit yyyy was defined more than one time
in a logical partition in CSS z. A logical partition cannot access more than one device in a subchannel set with the
same device number.

ICP362I • ICP364I
IOCP determines if a logical partition can access a device by testing the access lists of the chpids assigned to the
devices along with the device candidate lists (PART, PARTITION, or NOTPART keyword in the IODEVICE
statement). The duplicate device number error occurs if an access list of a chpid assigned to each device includes the
same logical partition and the device candidate lists of each device include the same logical partition and the devices
are in the same subchannel set.
Programmer response: Ensure that device numbers in a subchannel set are not duplicated within a logical partition.
Logical partitions must use unique device numbers in a subchannel set.
Do one of the following:
v Change the device number so that it is not duplicated
v Modify the initial access lists of the chpids assigned to the two devices so that they do not specify the same logical
partition
v Modify the device candidate lists so that they do not specify the same logical partition
ICP362I TOO MANY DEVICES ASSIGNED TO CONTROL UNIT xxxx

Explanation: You specified too many devices for control unit xxxx. You can define a maximum of 256 devices for a
control unit.
Programmer response: Reduce the number of devices attaching to control unit xxxx.
ICP363I CONTROL UNIT HEADER SPECIFICATION ERROR FOUND WHILE PROCESSING CONTROL
UNIT nnnn IN CSS x
Explanation: IOCP found a specification error while building a control unit header (CUH). A CUH can have a
maximum of eight paths in it. The CUH in CSS x containing control unit nnnn exceeds this maximum.
Programmer response: Either reduce the number of channel paths attaching to physical control units in the CUH or
reduce the number of physical control units in the CUH by changing the way devices are shared between the control
units.
ICP364I CHPID xx IN CSS w CONNECTS TO CHPID yy IN CSS z BUT CHPID yy IN CSS z [HAS NOT
BEEN DEFINED | IS NOT TYPE=nnn | IS NOT AN INTERNALLY CONNECTED TYPE=nnn
CHPID]
Explanation: You specified channel path yy in CSS z in the CPATH parameter on the CHPID statement for channel
path xx in CSS w. See the following explanations:
v HAS NOT BEEN DEFINED - channel path yy in CSS z must be defined in the I/O configuration and have the
same channel path type as channel path xx in CSS w
v IS NOT TYPE=nnn - channel path yy in CSS z does not have the same channel path type as channel path xx in CSS
w (that is, ICP, CIB, CL5, or CS5)
v IS NOT AN INTERNALLY CONNECTED TYPE=nnn CHPID - channel path xx in CSS w specifies a CSYSTEM
keyword value (CHPID statement) that matches the LSYSTEM keyword value (ID statement) for the I/O
configuration. Therefore, channel path xx is defined for an internal coupling connection. However, the connecting
channel path yy in CSS z has a different CSYSTEM keyword value. Both CSYSTEM keyword values must match
the LSYSTEM keyword value to establish an internal coupling connection.
Programmer response: Ensure that the following statements are true:
v Channel path yy in CSS z is defined
v Channel path yy in CSS z has the same channel path type as channel path xx in CSS w

ICP365I • ICP368I
v For internally connected CIB, CL5, and CS5 channel paths, channel path yy in CSS z must specify the same
CSYSTEM keyword value as channel path xx in CSS w that matches the LSYSTEM keyword value. If the coupling
connection is not internal to the CPC, change the CSYSTEM keyword value for channel path xx in CSS w to match
the name of the target CPC for channel path yy in CSS z.
ICP365I CHPID xx IN CSS m [ON CSYSTEM cccccccc] IS ALREADY CONNECTED TO CHPID yy IN CSS n
AND CANNOT CONNECT TO CHPID zz IN CSS w
Explanation: If the message is not for CSYSTEM cccccccc, you specified ICP, CIB, CL5, or CS5 channel path xx in
CSS m in the CPATH parameter on the CHPID statement for channel path yy in CSS n. However, channel path xx is
also in the CPATH parameter on the CHPID statement for channel path zz in CSS w. An ICP, CIB, CL5, or CS5
channel path (TYPE keyword) can only connect to one channel path.
If the message is for CSYSTEM cccccccc, you attempted to connect CIB, CL5, or CS5 channel paths yy in CSS n and zz
in CSS w to the same channel path xx in CSS m on system cccccccc. You cannot connect two CIB, CL5, or CS5 channel
paths to the same destination channel path. That is, you cannot specify the same CSYSTEM and CPATH keyword
values for two CIB, CL5, or CS5 channel paths.
Programmer response: If the message is not for CSYSTEM cccccccc, specify a different, available ICP, CIB, CL5, or
CS5 channel path in the CPATH keyword for one of the channel paths yy or zz. However, if channel path xx is CIB,
CL5, or CS5 and you want an external coupling connection, change the CSYSTEM keyword value for channel path xx
to match the name of the target CPC (LSYSTEM keyword value on the ID statement).
If the message is for CSYSTEM cccccccc, specify a different, available CIB, CL5, or CS5 channel path in the CPATH
keyword or a different system name in the CSYSTEM keyword for one of the channel paths yy or zz.
ICP366I CHPIDS xx AND yy IN CSS z HAVE INVALID CANDIDATE LISTS

Explanation: Channel paths xx and yy in CSS z are ICP, CIB, CL5, or CS5 channel paths (TYPE keyword) and are
connected to each other. However, both channel paths have candidate lists with the same, single logical partition (LP)
which is not allowed.
Programmer response: Specify multiple LPs in at least one of the candidate lists for the channel paths, or specify
different LPs in each candidate list. However, if the channel paths are CIB, CL5, or CS5 and you want external
coupling connections, change the CSYSTEM keyword value for each channel path to match the name of its target
CPC (LSYSTEM keyword value on the ID statement).
ICP367I CHPID xx IN CSS y CONNECTS TO ITSELF

Explanation: You specified ICP, CIB, CL5, or CS5 channel path xx in CSS y in the CPATH and PATH parameters on
a CHPID statement such that the channel path connects to itself. You must connect an ICP, CIB, CL5, or CS5 channel
path to another channel path and not to itself.
Programmer response: Connect channel path xx to another channel path. If the channel path is CIB, CL5, or CS5
and you want an external coupling connection, change the CSYSTEM keyword value for the channel path to match
the name of the target CPC (LSYSTEM keyword value on the ID statement).
ICP368I NONE OF THE DEVICES ASSIGNED TO CONTROL UNIT xxxx ACCESS CSS y
Explanation: You specified a null device candidate list for every device assigned to control unit xxxx such that no
device accesses CSS y of the control unit. If a control unit has devices, every CSS for the control unit must have at
least one device that can access it.
Programmer response: Either remove CSS y from the PATH keyword for the control unit or modify the device
candidate lists so that at least one device can access CSS y on the control unit.

ICP369I • ICP400I
ICP369I LSYSTEM KEYWORD IS REQUIRED ON THE ID STATEMENT WHEN TYPE=xxx CHPIDS ARE
DEFINED
Explanation: You defined CIB, CL5, or CS5 channel paths (TYPE keyword) but did not specify the LSYSTEM
keyword on the ID statement. The LSYSTEM keyword is required when CIB, CL5, or CS5 channel paths are defined.
Programmer response: Specify the LSYSTEM keyword on the ID statement.
ICP370I CHPIDS xx IN CSS m AND yy IN CSS n CONNECT TO EACH OTHER AND BOTH ARE
DEFINED ON PORT w OF [AID zz | PCHID zzz]
Explanation: You defined CIB, CL5, or CS5 channel paths (TYPE keyword) on the same port of an AID or PCHID
and connected them together with the CPATH keyword.
Programmer response: When connecting CIB, CL5, or CS5 channel paths within the same CPC, ensure that either
they are from different ports on the same AID or PCHID or that they have different AIDs or PCHIDs.
ICP371I MIXTYPE KEYWORD REQUIRED FOR PCHID xxx BECAUSE PCHID yyy IS DEFINED AS
TYPE=zzz
Explanation: You specified an invalid mix of channel path types for PCHIDs xxx and yyy. When defining FC and
FCP channel paths on the same FICON adapter on some CPCs, the MIXTYPE keyword is required. See Table 16 on
page 285 to determine which CPCs require the MIXTYPE keyword when defining FC and FCP channel paths on the
same adapter.
Programmer response: If the FICON adapter for the given PCHIDs supports the mixing of FC and FCP channel
paths, specify the MIXTYPE keyword on one or both of the CHPID statements for those PCHIDs. If the FICON
adapter does not support mixing of FC and FCP channel paths, change one of the channel path types so that the
PCHIDs have the same channel path type. See “MIXTYPE=” on page 70 for more information about the MIXTYPE
keyword on the CHPID statement and which FICON adapters support the mixing of channel path types.
ICP372I **CAUTION** PCHID xxx EXCEEDS THE MAXIMUM NUMBER OF PCHIDS ALLOWED FOR ITS
RESOURCE GROUP (SEE PCHIDS: yyy, zzz)
Explanation: You specified too many PCHIDs for a resource group. The PCHID slots in I/O drawers on some CPCs
are assigned to resource groups. Some CPCs restrict the number of ROCE and ROC2 function types that should exist
in a resource group. PCHIDs yyy and zzz are already defined in the same resource group and have reached the
maximum number allowed and therefore PCHID xxx exceeds that maximum. Exceeding the maximum can result in
some FIDs for these PCHIDs being unusable. To determine which PCHIDs and I/O slots are in the same resource
group for your CPC, see Table 3 on page 3. To determine the maximum number of ROCE and ROC2 PCHIDs your
CPC supports in a resource group, see Table 18 on page 287.
System action: IOCP prints this caution message and continues processing.
Programmer response: Move one of the ROCE or ROC2 PCHIDs to an I/O slot in another resource group.
ICP400I READ/WRITE OF IOCDS IS INVALID ON THIS PROCESSOR (xxxx)

Explanation: An option to print a report from an IOCDS or to write an IOCDS to the hard disk has been specified.
However, running ICPIOCP (z/OS or z/VSE) or specifying ICP for the IOCP command program option (VM) only
reads or writes an IOCDS for those CPCs that require the ICP version of IOCP. The CPC has a machine type of xxxx.
See Table 18 on page 287 to determine which CPCs ICP IOCP supports.
System action: IOCP sends message ICP051I and ends.
Programmer response: If you are verifying an input file that was written for those CPCs that require the ICP
version of IOCP, perform one of the following and run the job again:
v For z/OS and z/VSE, omit the WRTCDS and REPORT options in the PARM parameters of the EXEC statements.

ICP401I • ICP404I
v For z/VM, enter the ICP program option and omit the configuration report and write IOCDS options in the IOCP
command.
v If you want to read or write an IOCDS, you must use the IOCP required by your CPC.
Note: If your CPC supports ICP IOCP, you might not have the correct IOCP level for your CPC. You must use an
IOCP release level that supports your CPC. To help you determine the minimum level of IOCP required for your
CPC, see “Summary of changes” on page xxi.
Problem determination: Related message ICP051I indicates the function IOCP attempted.
ICP401I READ/WRITE IOCDS FACILITY TEMPORARILY BUSY (RC=x). RETRY LATER

Explanation: IOCP attempted to read or write an IOCDS. However, the facility that reads and writes IOCDSs is
temporarily busy. Return code 4 indicates that the is temporarily busy and could not process your read/write IOCDS
request. Return code 8 indicates that the operating system facility that handles read/write IOCDS requests is busy
with another request and could not process your request.
Programmer response: Wait a few minutes and rerun the IOCP. If the problem occurs again, notify the system
programmer.
ICP402I IOCP RECEIVED AN UNEXPECTED RESPONSE CODE OF xxxx HEX ATTEMPTING READ OR
WRITE OF AN IOCDS
Explanation: IOCP attempted to read or write an IOCDS and received an unexpected response code. You might
have received one of the following response codes:
v Response codes 0040 and 05F0 indicate that the read/write IOCDS facility of the is currently unavailable
v Response code 04F0 indicates that two users are attempting to write an IOCDS at the same time. Only one job
fails. The IOCDS written is associated with the successful job.
System action: IOCP ends the read or write operation.
Programmer response: For response codes 0040 and 05F0, reboot the support element, and then rerun the failing
job. Otherwise, wait a few minutes and rerun the failing job. If the problem occurs again, notify your service
representative.
Problem determination: If you are running IOCP on VM or as a guest on VM, see “IOCP command – usage notes
for VM” on page 125 for possible reasons for unexpected response codes.
ICP403I LEVEL Ax IOCDS IS INVALID

Explanation: IOCP received a response from the SERVICE CALL SVC (SVC 122) where the return code field in the
SERVICE CALL data block is 6A (hex). This message indicates that IOCP is trying to read an IOCDS that either is
being updated or was not closed after the last time data was written to it. The IOCDS is not considered valid.
System action: IOCP ends the read operation.
Programmer response: Rerun the job. If the problem occurs again, rewrite the IOCDS.
ICP404I LEVEL Ax IOCDS HAS INVALID FORMAT OR DIRECTORY

Explanation: IOCP has read part of an IOCDS into storage from the hard disk and compared the member names
and sector sizes in the directory record and the IOCDS format against the expected values. Either the IOCDS format
or directory does not match the expected values.
This situation occurs when the ICP IOCP tries to read an IOCDS that was written by a version of IOCP other than
ICP (for example, IYPIOCP for z/OS and VSE and program option IYP for VM).
Programmer response: Ensure that you have run the correct version of IOCP. Rerun the job. If the problem occurs
again, notify the system programmer.

ICP405I
ICP405I IOCP RECEIVED AN UNEXPECTED RETURN CODE OF xxxx HEX ATTEMPTING READ OR
WRITE OF AN IOCDS
Explanation: IOCP attempted to read or write an IOCDS and received an unexpected return code.
System action: IOCP ends the read or write operation.
Programmer response:
Code Description
09, 65 Either return code indicates that two users are attempting to read or write an IOCDS at the same time. Wait
a few minutes and rerun the job.
Note: If two jobs are attempting to write the same IOCDS, only one fails, but the IOCDS that is written
might be unusable.
50 You attempted to write an IOCDS to a remote CPC that is not defined to a Hardware Management Console
with changes enabled. Restart HCD and attempt the IOCDS build again.
51 You attempted to write an IOCDS to a remote CPC, and it did not respond. Ensure that the for the CPC is
powered on and rerun the job.
66 You attempted to read from an IOCDS or write to an IOCDS that is incorrect for your CPC. Specify a valid
IOCDS and rerun the job.
67 You attempted to write one or more write-protected IOCDSs. Remove the write protection and rerun the job.
68 You attempted to write an IOCDS that the CPC does not support (that is, writing an IOCDS with ICP IOCP
to a CPC that does not support ICP). If you attempted to write an IOCDS to a 2064 CPC in preparation for
upgrading it, see “Machine rules” on page 285 to determine which 2064 CPCs allow writing an IOCDS with
ICP IOCP.
6A You attempted to read from an IOCDS made incorrect by a write. If a write is currently in progress, wait
until the write has completed and rerun the job. If a write is not in progress, you must rewrite the IOCDS
before reading it.
80 You attempted to read an IOCDS that was never created. Specify an existing IOCDS and rerun the job.
82 You attempted to read an IOCDS from or write an IOCDS to the hard disk that is on the other side of a
physically partitioned multiprocessor; or you attempted to read an IOCDS from or write an IOCDS to the
hard disk for a nonexistent side of a non-multiprocessor.
83, 84 Another logical partition has a read or write in progress. Wait a few minutes and then rerun the job.
85 You are reading an IOCDS or IOCP source file made incorrect for one of the following reasons:
v A previous partition reset occurred while an IOCP write operation was in progress.
v IOCP was previously canceled while a write operation was in progress.
If the IOCDS is invalid and you cannot print a report of it or disassemble it, you must rewrite the IOCDS. If
the IOCP source file is invalid and you cannot build a new IOCDS, you must disassemble the current
IOCDS and allow the IOCP source file to be completely written. Then you can import your new IOCP
source file. You cannot correct the IOCP source file by only importing a file or using the Support Element
editor.
86 You attempted to read or write an IOCDS from an unauthorized logical partition. You can read or write an
IOCDS only from an authorized logical partition. Use the Change Logical Partition Security to authorize an
LP. For more information, see “Related information” on page xv for the appropriate operations guide.
Other return codes: If the return code is other than one listed in the preceding information, notify your service
representative.
Users of z/OS, z/VM, or z/VSE can see message ICP051I for more information about which IOCDS caused the
problem.

ICP406I • ICP409I
ICP406I READ/WRITE IOCDS REQUEST FAILED WITH ERROR CODE xxxx HEX
Explanation: IOCP attempted to read or write an IOCDS. However, the facility which reads and writes IOCDSs is
unavailable and failed with error code xxxx hexadecimal.
System action: IOCP sends message ICP0501I and terminates.
Programmer response: For z/VSE, ensure that VSE is not running as a guest under VM. VSE does not support
reading or writing an IOCDS as a guest under VM. Use the VM version of the IOCP.
If z/VSE is not running as a guest under VM, report the error code to your service representative.
For z/OS and z/VM, the read/write IOCDS facility of the CPC is not operational. Report the problem to your service
representative.
For z/VM, if the service representative does not find any problems with the CPC, the problem might exist with the
CMS-IOCP interface module (DMSICP).
ICP407I TARGET PROCESSOR xxxx-yyy DOES NOT MATCH REQUESTED PROCESSOR

Explanation: The Hardware Configuration Definition (HCD) requested that IOCP write an IOCDS to a target CPC at
a specific SNA address. Machine type xxxx and model number yyy are associated with the target CPC on which
IOCP attempted to write the IOCDS. The machine type and model number of the target CPC are not compatible with
the machine type and model number of the CPC you requested in HCD. One of the following conditions exist:
v You defined the CPC in HCD with the wrong machine type or model number.
v The SNA address defined for the CPC in HCD is incorrect.
System action: IOCP ends the write operation. If the target IOCDS has a status of Invalid, it is incorrect and
unusable. Otherwise, the IOCDS is unchanged.
Programmer response: Return to HCD, and do one of the following actions:
v Change the machine type and model number for the CPC to the correct values.
v Correct the SNA address for the requested CPC.
Attempt to build the IOCDS again.

Problem determination: Related message ICP051I indicates the function IOCP performed. The ID statement at the
beginning of the IOCP output listing contains information about the requested CPC.
ICP408I IOCDS Ax IS WRITE PROTECTED

Explanation: IOCP tried to write an IOCDS to the , but the IOCDS is write-protected.
System action: IOCP ends the write operation.
Programmer response: Remove the write protection and rerun the job. To remove the write protection, use Disable
Write Protection under Options on the .
ICP409I WRITE OF IOCDS IS INVALID BECAUSE IOCP CANNOT DETERMINE PROCESSOR MODEL
NUMBER (RC=xx)
Explanation: IOCP is unable to write the IOCDS because it cannot determine the model number of the CPC. Your
operating system might be missing maintenance that IOCP requires. IOCP uses the STSI machine instruction or
operating system services to determine the CPC model number.
Programmer response: Notify the service representative.

ICP410I • ICP502A
ICP410I IOCP WAS UNABLE TO OBTAIN SUFFICIENT STORAGE TO REPORT ON SYSTEM=(xxxx,n)

IOCDS Ay
Explanation: IOCP could not obtain enough storage by using the GETMAIN statement (for z/OS and z/VM) or the
GETVIS statement (for z/VSE) to report on IOCDS Ay associated with SYSTEM=(xxxx,n). The SYSTEM parameter
associated with the IOCDS might be different from the one associated with your CPC. IOCP uses a variable amount
of storage based on the type of IOCDS it is processing (that is, the SYSTEM parameter). To determine the specific
IOCP storage requirements for your CPC, see Table 18 on page 287.
System action: IOCP ends the read operation for the IOCDS. If you are reporting on multiple IOCDSs, IOCP
continues with the next IOCDS.
Programmer response: Increase the amount of storage available for IOCP use.
For z/OS users, see “IOCP storage requirements” on page 112.
For z/VM users, define more virtual storage, perform an IPL of CMS, and issue the IOCP command again. For more
information, see “IOCP storage requirements” on page 131.
For z/VSE users, increase the size of your partition and specify SIZE=AUTO on the EXEC statement of the job to
allow IOCP to use the remaining partition size for GETVIS. For more information, see “IOCP storage requirements”
on page 131.
ICP412I IOCDS WAS WRITTEN WITH IOCP VERSION nn.nn.nnn AND IS TOO LARGE TO BE READ BY
CURRENT VERSION OF IOCP
Explanation: IOCP has read part of an IOCDS into storage from the hard disk and compared the IOCDS member
sizes in the directory record with the maximum sizes supported by this version of IOCP and determined that the
IOCDS is too large to be read. This situation can occur when an IOCDS is written by a later version of IOCP than the
one reading it. The IOCDS being read was written with IOCP version nn.nn.nnn.
Programmer response: Read the IOCDS using the same version of IOCP that wrote it.
ICP500A PROCESSING COMPLETED. CHECK PRINTER FOR LISTING OR RETURN TO EDITOR.

Explanation: IOCP has processed the data entered on the display frame, completed the requested function, and
detected no errors. An output list was produced.
System action: IOCP waits for an operator response.
Operator response: Enter a command.
ICP501A PROCESSING COMPLETED. NO PRINTER SPECIFIED-NO LISTING. PROCEED.

Explanation: IOCP has completed the requested function, and detected no errors. An output list was not produced.
ICP502A PROCESSING COMPLETED BUT HAS MESSAGES. RETURN TO EDITOR.

detected a warning or caution message (no errors).
Operator response: Return to the editor to inspect the warning or caution messages and make any necessary
changes.

ICP503A • ICP507A
ICP503A PROCESSING COMPLETED. PROCEED.

detected no errors.
ICP504A IOCS IS TOO LARGE FOR IOCP TO READ

Explanation: The card-image file on the Support Element hard disk contains too many records for IOCP to read.
IOCP can read a maximum of 256020 cards. Twenty records are reserved for IOCP messages.
Operator response: You can reduce the size of the card-image file by doing one or more of the following actions:
v Remove comment cards from the card-image file
If the card-image file size cannot be sufficiently reduced, contact your service representative or perform the IOCP
function from z/OS, z/VM, or z/VSE if possible.
ICP505A IOCS IS TOO LARGE FOR IOCP TO WRITE

Explanation: The card-image file that IOCP read from the Support Element hard disk or disassembled from an
IOCDS contains too many records for IOCP to write to the Support Element hard disk. The maximum size of a
card-image file is 256020 cards, and 20 records are reserved for IOCP messages.
Operator response: For IOCP function Build data set, you can reduce the size of the card-image file by doing one or
more of the following actions:
v Remove comment cards from the IOCS.
v Consolidate IOCP statements that are continued across multiple cards onto a minimum number of cards.
If the card-image file size cannot be sufficiently reduced, contact your service representative or perform the IOCP
function from z/OS, z/VM, or z/VSE if possible.
For IOCP function Disassemble data set, contact your service representative.
ICP506A CARD INPUT ERROR(S) FOUND, RETURN TO EDITOR

Explanation: IOCP has been requested to Build an IOCDS. However, IOCP has detected that the card-image file
contains one or more errors.
Operator response: Return to the editor to inspect the error messages, correct the errors, and try again.
ICP507A IOCDS GENERATION ERROR, RETURN TO EDITOR

Explanation: IOCP has been requested to Build an IOCDS. However, IOCP has detected an error while trying to
generate an IOCDS.
Operator response: Return to the editor to inspect the error message, correct the error, and try again.

ICP508A • ICP550A
ICP508A CARD IMAGE FILE EXCEEDS ALLOCATED STORAGE

Explanation: IOCP has been requested to Build or Disassemble an IOCDS. However, IOCP does not contain enough
storage for the card-image file.
Operator response: Select a logical partition with more storage allocated and perform the function again.
ICP509I ENTRIES BEING PROCESSED. PLEASE WAIT FOR COMPLETION.

Explanation: This message indicates that IOCP is processing the entries for the requested function.
System action: IOCP locks the keyboard until the function is complete.
Operator response: Wait for and respond to the next message.
ICP510I ENTRIES BEING PROCESSED. LISTING WILL BE PRINTED.

Explanation: IOCP is processing the requested function and the function includes producing an output list.
ICP511I ENTRIES BEING PROCESSED. PRINTER NOT SPECIFIED.

Explanation: IOCP is processing the entries for the requested function, and, because a printer was not specified, the
function does not include an output list.
ICP512A LOGICAL PARTITION HAS INSUFFICIENT STORAGE

Explanation: You selected a logical partition with insufficient storage to perform the selected IOCP function.
Operator response: Select a logical partition with more storage allocated, and perform the function again. The
Support Element enforces the correct minimum storage requirement for stand-alone IOCP. Contact your service
representative.
ICP515W IOCP TERMINATED. WAIT STATE CODE = B2x

Explanation: IOCP has loaded the PSW with a wait state code of B20 through B24. See “IOCP wait state codes” on
page 148for an explanation of the wait state codes. Also, for more information, see “Related information” on page xv
for the appropriate stand-alone IOCP document.
System action: IOCP ends.
Operator response: See “IOCP wait state codes” on page 148.
Problem determination: See “IOCP wait state codes” on page 148.
ICP550A dddd, INT REQ, CC=3/NO PATH AVAILABLE

Explanation: Before reading from an input device or writing to an output device, IOCP received a not-operational
condition code (code 3) on the path specified. In the message, dddd indicates the device address or number.
System action: IOCP waits for the operator to:
1. Correct the entry or make the path available.
2. Try the operation again.
Operator response:
1. If the entry for the device was incorrectly specified on the frame, correct the entry and try the operation again.

ICP551A • ICP552A
2. If the path is not available, turn on the path to the device (such as setting the channel or control unit switch) and
try the operation again.
3. If the path cannot be made available, run IOCP again using another device or path.
ICP551A dddd,INT REQ,cm,stat,sense

Explanation: Before reading from an input device or writing to an output device, IOCP detected a device that
requires operator intervention.
In the message text, in hexadecimal, the fields are:
dddd The device address or number.
INT REQ
Intervention required.
cm The operation code of the channel command word (CCW) during whose execution the error occurred. If the
channel command word cannot be found, this field is shown as **.
stat The status portion of the channel status word (CSW).
sense The sense data can be a maximum of 23 bytes long. The first two sense bytes are for the error condition.
System action: IOCP waits for the operator either to end IOCP or to try the operation again.
Operator response: Do the following steps, and try the operation again:
v Make the unit ready. If the unit cannot be made ready, run IOCP again using another device.
v Place paper in the paper supply area.
v Remove paper from the output area.
ICP552A dddd,err,cm,stat,sense
Explanation: Before reading from an input device or writing to an output device, IOCP detected an I/O error that
cannot be corrected. Two consecutive commas or a blank field in the message text indicates that a field could not be
determined.
In the message text, the fields are:
dddd The device address or number, in hexadecimal.
err A description of the error based on status and sense information:
BOC Bus-out check.
CCC Channel control check.
CDC Channel data check.
CHC Chaining check.
CMD Command reject.
CPC Channel program check.
DCK Data check.
EQC Equipment check.
ICC Interface control check.
IOE Input/output error (for errors other than those errors described).
OVR Data overrun.
PRC Channel protection check.
SEN A unit check occurred during a sense operation. (When this condition is present, the sense field is
not included in the message text.)
cm The command code, in hexadecimal, of the channel command word (CCW) being processed when the error
occurred. If the channel command word cannot be found, this field is shown as **.

ICP553A • ICP554A
stat The status portion, in hexadecimal, of the channel status word (CSW).
System action: IOCP waits for the operator either to try the operation again or to end IOCP.
Operator response: This error is probably a user or hardware error. Try another device on another channel.
Possible values of the err field and appropriate responses are as follows:
CMD REJECT
Command reject. Check that device entries on the frame are correct.
BOC Bus-out check.
OVR Data overrun.
These are permanent hardware faults. Service representative action is required.
ICP553A dddd, INT REQ, CC=3/NO PATH AVAILABLE

Explanation: While reading from an input device or writing to an output device, IOCP received a not-operational
condition code (code 3) on the path specified. In the message, dddd indicates the device address or number.
System action: IOCP waits for the operator to make the path available and to signal IOCP by pressing Enter.
Operator response:
1. If the path is not available, turn on the path to the device (such as setting the channel or control unit switch) and
try the operation again.
2. If the path cannot be made available, run IOCP again using another path or device.
ICP554A dddd,INT,REQ,cm,stat,sense
Explanation: While reading from an input device or writing to an output device, IOCP detected that the device
requires operator intervention.
In the message text, in hexadecimal, the fields are:
dddd The device address or number.
INT REQ
Intervention required.
cm The operation code of the channel command word (CCW) during whose execution the error occurred. If the
channel command word cannot be found, this field is shown as **.
stat The status portion of the channel status word (CSW).
System action: IOCP waits for the operator to press Enter.
Operator response: Do the following steps:
v Make the unit ready. If the unit cannot be made ready, run IOCP again using another device.
v Place paper in the paper supply area.
v Remove paper from the output area.
Press Enter.

ICP555A • ICP556I
ICP555A dddd,err,cm,stat,sense
Explanation: While reading from an input device or writing to an output device, IOCP detected an error that can be
tried again or ignored. Two consecutive commas or a blank field in the message text indicates that the field could not
be determined.
DCK Data check
EQC Equipment check (card reader)
System action: IOCP waits for the operator to signal IOCP to try the operation again by pressing Enter. If the error
occurred on the input device, IOCP tries the read operation again. If the error occurred on the output device, IOCP
does not reprint the line.
Operator response: Have IOCP try the command again (by pressing Enter), or rerun IOCP using another device.
ICP556I dddd,err,cm,stat,sense
Explanation: While IOCP was reading from an input device or writing to an output device, IOCP detected an error
that cannot be corrected. Two consecutive commas or a blank field indicates that a field could not be determined.
BOC Bus-out check.
CCC Channel control check.
CDC Channel data check.
CHC Chaining check.
CMD Command reject.
CPC Channel program check.
IOE Input/output error (for errors other than those errors described).
OVR Data overrun.
PRC Channel protection check.
REC Record error. (The record read was not 80 bytes or a multiple of 80 bytes.)
SEN A unit check occurred during a sense operation. (When this condition is present, the sense field is
not in the message text.)

DMSICP001E • DMSICP002E
System action: IOCP ends by loading a PSW with a wait state code of B23. This error is probably a hardware error.
(Try another device on another channel.)
For a magnetic tape unit, do not unconditionally accept the results of the operation. Check that the tape being used
for this job has not been replaced or removed. Consider each message as a potential warning of a marginal condition.
For a card reader, some abnormal error condition occurred. Depending on the severity of the error (check the status
information and the sense information) and, depending on the site requirements, take the appropriate action.
Possible values of the err field and appropriate responses are as follows:
CMD REJECT
Command reject. Correct the command and try the function again.
BOC Bus-out check.
OVR Data overrun.
These are permanent hardware faults. Service representative action is required.
IOCP messages (DMSICP prefix)

Only the VM version of IOCP sends the following messages. The messages are arranged in alphanumeric
order by message identifier.
The message format is:

DMSICPnnnE text
nnn Is the message number.
text Is the message text.
Note: The CMS-IOCP interface module sends these error messages with corresponding return codes
(RCs) for each message.
DMSICP001E No filename specified

Explanation: The IOCP command requires you to specify the name of the file containing the source IOCP
statements or the name of the file to contain the IOCP output file.
System action: Processing of the command ends. The system status remains the same. RC=24.
User response: Issue the command again and specify the file name of the IOCP input or output file.
DMSICP002E File 'filename' not found

Explanation: The specified file (file type IOCP) was not found on any accessed disk. Either the file is not on any
accessed disk or the file name was misspelled.
User response: Find or create the wanted file. To make sure that the file exists, use either one of the following
commands:
STATE filename IOCP *
or
LISTFILE filename IOCP *
Correct the command and issue it again.

DMSICP003E • DMSICP075E
DMSICP003E Invalid option: 'option'

Explanation: The specified option is incorrect.
v The option might have been misspelled.
v If the option can be truncated, it might have been truncated improperly.
v The option might conflict with another option in the command line.
System action: Processing of the command is ended. The system status remains the same. RC=24.
User response: Correct the command and issue it again.
DMSICP006E No read/write disk accessed

Explanation: You do not have access to a read/write disk on which the command can write its output file.
User response: Access a read/write disk and issue the command again, or use the CP LINK command to reset the
A-disk to read/write mode. Access the A-disk again, and issue the command.
DMSICP007E File 'filename' IOCP is not fixed, 80-character records

Explanation: The specified file must have fixed-length, 80-character records for the command to be processed.
User response: It is possible that an incorrect file identification was specified in the command line. In this situation,
issue the command again. If, however, the file identification was correct but the file has the wrong format or does not
contain 80-character records, change the format or record length of the file with the COPYFILE or XEDIT command.
DMSICP038E Fileid conflict for DDNAME SYSIN

Explanation: You issued the IOCP command for a file (file type IOCP) that exists on a disk. Before this command,
you sent a FILEDEF command that does not point to the same file. The CMS-IOCP interface module does not choose
between the two valid file definitions and sends this error instead.
User response: Check that you specified the correct file name with the IOCP command. Then, do one of the
following actions:
v If the file name is correct, issue a FILEDEF SYSIN CLEAR command for the file.
v If the file name is not correct, use a different file name on the IOCP command or erase the existing disk file (file
type IOCP).
DMSICP070E Invalid parameter 'parameter'

Explanation: An incorrect operand was specified in the command line.
User response: Correct the command line and issue the command again.
DMSICP075E Device 'devtyp' invalid for input

Explanation: The device specified for the input ddname (SYSIN) is incorrect. This message is displayed if the input
device specified is DUMMY, PRINTER, PUNCH, TERMINAL, or GRAF ('devtyp' = CRT).
User response: Issue the FILEDEF command again and specify the correct input device.

DMSICP099E
DMSICP099E CMS/DOS environment active

Explanation: The IOCP command cannot be processed while the CMS/DOS environment is active.
User response: Use the SET DOS command to deactivate the CMS/DOS environment. Then issue the IOCP
command again.

Appendix A. Coding IOCP statements
This appendix describes the rules for coding IOCP card-image statements and the notation this document
uses to describe the statements.
Rules for coding IOCP statements

The rules for coding IOCP statements are the rules of the assembly language. The following paragraphs
are a summary of these rules as stated in High Level Assembler for MVS & VM & VSE Language Reference,
SC26-4940.
IOCP statements have the following standard format:

Name
Operation
Operand
Comments
Name Symbolically identifies the statement. If included, it must contain 1-8 alphanumeric or special (#,
@, or $) characters. The first character must be an alphabetic or special character. Only uppercase
characters are allowed. The name must begin in the first column of the statement and must be
followed by one or more blanks. IOCP ignores the name field of an IOCP statement.
Operation
Identifies the statement. It must be preceded and followed by one or more blanks. The operation
field can start in the second column of the statement if the name field is not used.
Operand
Contains parameters coded in any order and separated by commas. The operand field ends with
one or more blanks placed after the last parameter. However, for the ID statement, IOCP permits
a comma to be coded as the last character of the operand field.
Most parameters consist of a keyword followed by an equal sign (=) and the keyword value. The
keyword value can be a single value (or subfield) or a list of values (or subfields). If the keyword
value consists of more than one subfield, the subfields must be separated by commas and the list
of subfields must be enclosed in parentheses.
When a subfield contains multiple values (such as the UNITADD= subfield in the CNTLUNIT
statement), these subparameters are positional and must be coded in the order shown. The
operand field can contain a maximum of 1024 characters.
Comments
Can be written in an IOCP statement, but they must be separated from the last parameter of the
operand field by one or more blanks. You can use an entire card for a comment by placing an
asterisk (*) in the first column or a period and asterisk (.*) in columns 1 and 2. A maximum of
two continuation cards can be used for comments.
Note: Stand-alone IOCP removes all comments that begin with *ICP and *IYP. Stand-alone IOCP uses
*ICP for IOCP messages inserted into the card-image file. By removing all preexisting *ICP comments,
stand-alone IOCP ensures that all *ICP comments in the card-image file after IOCP processing are only
IOCP messages.
For HCD: HCD might generate comments that begin with *$HCDC$ and *$HCD$. They have special
meaning to HCD; see the OS/390 Hardware Configuration Definition User's Guide, SC28-1848, z/OS Hardware
Configuration Definition User's Guide, SC33-7988, or z/VM Dynamic I/O Configuration Planning and
Administration, SC24-6044.

IOCP statements are coded in columns 1-71 of each card. You can continue a statement that exceeds 71
columns onto one or more additional cards by placing a nonblank character in column 72 to indicate the
continuation. The statement can be interrupted either at column 71 or after any comma that separates
parameters. The continued portion must begin in column 16 of the following card.
Continued cards can include comments. You can use columns 73-80 to code identification or sequence
characters if you choose. IOCP prints but does not examine columns 73-80.
Note: If you incorrectly continue a statement and only optional parameters are present on the continued
portion, IOCP ignores the optional parameters. For all statements except ID, IOCP issues a message
following a continuation card that is treated as a comment by IOCP and is ignored. If there are no
parameters but a valid comment exists on the continuation card, IOCP still issues the message. Placing
the comment on a comment card instead of a statement card removes the message.
For example, if you do not end the last parameter on a card with a comma but end it with a blank and
code a nonblank character in column 72, IOCP ignores the information about the continuation card. Also,
if you end the last parameter on a card with a comma and code a nonblank character in column 72 but
begin the continuation card in column 17 or higher, IOCP ignores the information about the continuation
card.
To avoid this situation, you can code the optional parameters on the first card of the statement. Check the
I/O configuration reports to ensure that all channel paths, control units, and I/O devices are defined
correctly.
Format and coding conventions

The conventions in this publication that illustrate the format and coding of IOCP statements are:
v You must code uppercase letters, numbers, and punctuation marks exactly as shown.
Exceptions to this convention are brackets ([ ]), braces ({ }), and ellipses (...). You never code these
characters.
v Lowercase letters represent variables for which you must substitute information or specific values.
v Items enclosed in braces ({ }) represent alternative items. You must code only one of the items.
v Items enclosed in brackets ([ ]) are optional. You can omit them. Conversely, the lack of brackets
indicates that an item must be coded.
v An ellipsis (...) indicates that the previous item or group of items can be coded two or more times in
succession.
v The "or" sign (|) separates alternative items.
v An alternative item that is underlined is the default value. IOCP uses the default value if you do not
specify the keyword.
v You must enclose single-value subfields in parentheses if you code more than one of them. If you code
only one subfield, you can omit the parentheses. For example, you can code either CUNUMBR=(530) or
CUNUMBR=530 in the IODEVICE statement.
v You must enclose the UNITADD= subfields in the CNTLUNIT statement in double parentheses
because these subfields can contain multiple values. (Single parentheses enclose each subfield; another
set encloses all the subfields.) For example, you code UNITADD=((00,16),(20,8)). If you code only a
single subparameter for one subfield, you can omit the parentheses. For example, you can code either
UNITADD=((0A)) or UNITADD=0A.
v Parameters, subfields, and subparameters coded in the operand field must be separated by commas.
Example
A typical statement might be:
CU10A CNTLUNIT CUNUMBR=10A,PATH=06,SHARED=N, X
UNIT=2821,UNITADD=((0A,3)) Plan 3

CU10A is the symbolic name of the statement.
CNTLUNIT identifies the statement to the system.
CUNUMBR=10A, PATH=06, and SHARED=N are required parameters, separated by commas, containing
keywords and keyword values. Because the statement is continued, a comma follows SHARED=N, and a
nonblank character (X) is placed in column 72.
UNIT=2821, and UNITADD=((0A,3)) are also required parameters and start in column 16 of the continued
statement. Because UNITADD=((0A,3)) is the last parameter, it is followed by a blank to indicate the end of
the operand field.
Plan 3 is a comment.
Appendix A. Coding IOCP statements 239

Appendix B. Lists of statement input
Figure 31 shows a listing of a sample IOCP input file.
ID MSG1=’SAMPLE IOCP INPUT’

*
RESOURCE PARTITION=(CSS(0),LP1,LP2,LP3)
*
CHPID PCHID=100,PATH=20,TYPE=CVC,PART=(LP1,REC)
CHPID PCHID=101,PATH=21,TYPE=CNC,SHARED
*
CNTLUNIT CUNUMBR=000,PATH=(20),PROTOCL=D,SHARED=N,UNIT=3274, +
UNITADD=((C0,32))
IODEVICE ADDRESS=(0C0,32),CUNUMBR=(000),UNIT=3279,MODEL=3B
*
CNTLUNIT CUNUMBR=010,PATH=(21),UNIT=3990,UNITADD=((00,32))
IODEVICE ADDRESS=(100,32),CUNUMBR=(010,011),UNIT=3390
*
IODEVICE ADDRESS=(480,16),CUNUMBR=(020,021),UNIT=3480,UNITADD=00
Figure 31. Sample IOCP statements
Figure 32 shows a listing of sample SYSTEM CONFIG statements used for VM I/O definition.
/***************************************************/
/* RDEVICE statements */
/***************************************************/
Rdevice 100-108 Type 3270_DISPLAY
Rdevice 181 Type CTCA
Rdevice 182 Type 3420 Model 4
Rdevice 190 Type Tape
Rdevice 3C0-3C7 Type DASD
Rdevice 9000 Type IMPACT_PRINTER NOTSENSED
Figure 32. Sample SYSTEM CONFIG statements for VM I/O definition
Figure 33 on page 242 shows a listing of an HCPRIO file used for VM system generation.

FILE: HCPRIO ASSEMBLE A SP/CMS REL 2.1 CSL 24 DSD - COMSYS PAGE 001
RIO TITLE ’HCPRIO -- SAMPLE’ HCP00010

RDEVICE ADDRESS=001,DEVTYPE=3211,CLASS=(A),FEATURE=UNVCHSET HCP00020
RDEVICE ADDRESS=012,DEVTYPE=3505 HCP00030
RDEVICE ADDRESS=013,DEVTYPE=3525,CLASS=C HCP00040
RDEVICE ADDRESS=(020,16),DEVTYPE=3279,MODEL=3 HCP00050
RDEVICE ADDRESS=090,DEVTYPE=3705,ADAPTER=BSCA,BASEADD=0F0 HCP00060
RDEVICE ADDRESS=0F0,DEVTYPE=3705,ADAPTER=TYPE1,MODEL=F8 HCP00070
RDEVICE ADDRESS=0F1,DEVTYPE=3705,ADAPTER=TYPE4,MODEL=F8 HCP00080
RDEVICE ADDRESS=(140,8),DEVTYPE=3350,CUTYPE=3880-11 HCP00090
RDEVICE ADDRESS=(168,8),DEVTYPE=3350 HCP00100
RDEVICE ADDRESS=(3B0,8),DEVTYPE=3380 HCP00110
RDEVICE ADDRESS=(3C0,8),DEVTYPE=3380 HCP00120
RDEVICE ADDRESS=(3D0,8),DEVTYPE=3380 HCP00130
RDEVICE ADDRESS=(11A0,16),DEVTYPE=3380 HCP00140
RIOGEN CONS=020,ALTCONS=021 HCP00150
END HCP00160
Figure 33. HCPRIO file for VM system generation
Figure 34 shows the VSE system generation input that corresponds to the IOCP statements in Figure 31
on page 241.
Note: See Chapter 7, “IOCP configuration reports,” on page 151 for examples of the reports IOCP
produces.
ADD 0C0:0DF,3279
ADD 100:11F,3390
ADD 480:48F,3480
Figure 34. Source input for VSE system generation

Appendix C. Characteristics of the I/O interface timeout
function
In the CPC, each channel path has I/O interface timeout functions that time the control unit delays in
completing the following I/O interface sequences:
v A 4-second timeout for all selection and status presentation sequences. A timeout occurs if the sequence
is not complete within 4 seconds.
v A 30-second timeout for data transfer. A timeout occurs if a byte of data is not transferred within 30
seconds.
If a timeout occurs, the channel terminates the I/O request to the control unit and generates an
interface-control-check interruption.
For additional information about these sequences, see System/360 and System/370 I/O Interface Channel to
Control Unit Original Equipment Manufacturers' Information, GA22-6974.
Malfunctions in control units and I/O devices can cause a channel path to be unavailable to other control
units and I/O devices. The timeout function detects these malfunctions.
When the timeout function is inactive (timeout is inhibited), control unit delays are not timed and the
channel unconditionally waits until the control unit completes the sequence or initiates the next tag
sequence. If the control unit fails to complete a sequence because of a malfunction, the channel path
remains active to the control unit. In this situation, the channel path is unavailable for I/O requests to
other control units until the failing control unit deactivates all in-tags (except metering-in and request-in)
or until the channel path is reset.
The channel times I/O requests as follows:

v If TIMEOUT=Y is specified, the I/O interface timeout function is active and the channel times all I/O tag
sequences and delays between dependent tag sequences that are initiated by either the channel or the
control unit. Some examples are:
– The initial selection sequence resulting from a Start I/O instruction or a Start I/O Fast Release
instruction
– A request-in tag sequence initiated by the control unit to reconnect to the I/O interface
– Delays between data transfer tag sequences
v If TIMEOUT=N is specified, the I/O interface timeout function is inactive (timeout is inhibited) for:
– Data transfer tag sequence
– Delays between data transfer tag sequences
Timeout is inhibited from the rise of service-out that follows status-in during the initial selection sequence
until the rise of service-out that follows status-in during the ending sequence.
Examples follow that show when the timeout function is inactive if TIMEOUT=N is specified.
Example A. Channel initiated sequence (for example, SIO)

Initial Data Ending status:
selection transfer channel end and
sequence sequences device end sequence
<---Timed-----><---Not timed-----><--------Timed------------->
Example B. Command chaining control unit initiated reconnection sequence

Device end Initial Data Ending status:
reconnect selection transfer channel end and
sequence sequence sequences device end sequence
<--------Timed------------><--Not timed--><---------Timed----------->
When a timeout condition is detected, the channel terminates the I/O request to the control unit by
issuing a selective reset to the control unit and then generating an interface-control-check interruption. A
channel report word (CRW) is made pending, which indicates a channel-path-terminal condition.
If TIMEOUT=N is specified, detection of a timeout condition as a result of an interface associated

malfunction might not occur (see the beginning of this appendix for a description of sequences that are
not timed when TIMEOUT=N is specified). Failure to detect and recover this condition might make various
resources unavailable to the control program. If continued operation is dependent on the availability of
these resources, system degradation might be experienced until those resources are recovered. Recovery
might occur as a consequence of detecting a secondary error the CPC might develop from failing to
detect the timeout condition. Generally, this situation does not occur and recovery of the resources
requires operator intervention. For these reasons, code TIMEOUT=N only when you decide a real need exists
that justifies the exposure.
Timeout considerations
Some control units and I/O devices have valid operational conditions that exceed the 30-second I/O
interface timeout function limit. If timeout is active when such a condition occurs, the channel terminates
the I/O request and generates an interface-control-check interruption, even though no malfunction exists.
For example:
v A timeout might occur during a read operation to a magnetic tape unit if the tape being read is blank.
While the magnetic tape unit searches for a data byte or the end-of-tape mark, it might exceed the
30-second timeout function limit.
v A channel-to-channel adapter between two CPCs might time out if one system attempts to
communicate with the other system while the other CPC is in a stopped state.
When timeout is active, the channel might terminate an I/O request and generate an
interface-control-check interruption even though no control unit or I/O device malfunction has occurred.
Note: For a device with valid operational conditions that can exceed the 30-second I/O interface timeout
limit, inhibit the timeout function by specifying TIMEOUT=N in the IODEVICE statement. For other IBM
I/O devices, specify the default TIMEOUT=Y.

Appendix D. List of I/O devices and control units
Usage notes:
The list does not completely describe every available control unit and device for CPCs that support the
IOCP that this publication describes.
This document does not define all device requirements. It is meant to be a general reference guide for
configuring nonspecific I/O devices. For information concerning the configuration of specific I/O devices,
consult the device planning manuals associated with those devices. They contain the most current,
accurate information concerning device configuration. Appendix D provides some information for some
specific I/O devices, but do not consider this appendix the main or complete source for such information.
For the CPC, the tables in this chapter provide a list of I/O devices and show the keyword values you
specify for the UNIT, MODEL, and STADET keywords in the IODEVICE statement. Also, they list the
control units that attach to the I/O devices and show the keyword values you specify for the UNIT,
SHARED, and PROTOCL keywords in the CNTLUNIT statement. Notes, which are referenced in the list
of devices and control units, contain additional information for specific I/O devices and control units.
See “Specifying IOCP statements” on page 53 for a full description of the IODEVICE and CNTLUNIT
statements and the meanings of the keyword values shown in the tables in this chapter.
The following items are presented for your review of I/O devices and control units:
1. IOCP might support devices that your operating system might not support.
2. Your operating system might support devices that the IOCP described in this publication does not
necessarily support.
3. If you use the UNITADD keyword in the IODEVICE statement, you must follow the same rules and
restrictions that apply to the last two digits of the ADDRESS parameter.
4. If you specify the UNITADD parameter in the IODEVICE statement, the last two digits of the
ADDRESS parameter can be different from the physical unit address of the device if you follow these
rules:
v All device rules for addressing must be followed.
v For tape units and display devices, all unit addresses on a control unit must be consecutive and the
lowest unit address must be modulo2 to the number of devices that can be attached to the control
unit.
v The lowest unit address on a string of DASDs must be modulo to the maximum string size.
Table 14. IOCP parameter values for IODEVICE and CNTLUNIT statements
IODEVICE statement CNTLUNIT statement
UNIT= MODEL= STADET= Notes UNIT= Model SHARED= PROTOCL= Notes
Direct access storage devices
3330 1, 2, 11 Y 1 3880 1, 2, 11 N D or S 2, 3
3330V Y 4 3830 3 N D 4
3333 1, 11 Y 3880 1, 2, 11 N D or S 2, 3
3350 Y 1,5 3830 3 N D
2. Modulo means that the unit address must be evenly divisible by the number of devices in hexadecimal. For example, for 16
devices, the last digit of the unit address must be 0. For four devices, the last digit must be 0, 4, 8, or C.

Table 14. IOCP parameter values for IODEVICE and CNTLUNIT statements (continued)
3880 1, 2, 11 N D or S 2, 3
3350P Y 1,5 3880 11 N D or S 2, 3, 6
3351P Y 1,7 3880 21 N D or S 2, 3, 8
3375 Y 3880 1, 2 N D or S 2, 3
3380 Y 3880 2, 3 N D or S 2, 3
3880 13 N S 2
3880 21, 23 N S or S4 2
3990 1 N S 2, 3
3990 2 N S or S4 2, 28
3990 3 N S4 2, 28
49 2105 All 28, 50
3390 Y 24 3990 2,3,6 N S4 2, 28
49 2105 All 28, 50
3851 Y 3851 All N D
9345 Y 9341 N S or S4 31
9343 All N S or S4 30, 31
Display devices
3178 Y 3174, 3274 N or Y D 11, 36
3179 Y 3174, 3274 N or Y D 11, 36
3180 Y 3174, 3274 N or Y D 11, 36
3251 Y 3258 Y D
(by 3255) 5088 N D or S 12
3270 X 29,42 OSC
3270 PC Y 3174, 3274 N or Y D 11, 36
3277 Y 3272 N or Y D 11
3174, 3274 N or Y D 11, 36
3278 3174, 3274 N or Y D 11, 36
3279 3174, 3274 N or Y D 11, 36
3290 3174, 3274 N or Y D 11, 36
5081 (by Y 9 5088 N D or S 12
5085)
6580 (with Y 3174, 3274 N or Y D 11, 36
3270
feature)
PC (with Y 3174, 3274 N or Y D 11, 36
3277
feature)
Printers
1403 N1, 2, 7 Y 2821 1, 2, 3, 5 N D
3170 Y 3170 N D, S, or S4

3203 5 Y 17 3203 4 N D
3211 1 Y 18 3811 1 N D
3230 2 Y 3174, 3274 N or Y D 11
3262 3, 13 Y 3272 N or Y D 10
3174, 3274 N or Y D 11
3262 5 Y 3262 N D
3268 2 Y 3174, 3274 N or Y D 11
3286 Y 3272 N or Y D 10
3174, 3274 N or Y D 11
3287 Y 3272 N or Y D 10
3174, 3274 N or Y D 11
3289 2 Y 3272 N or Y D 10
Y 3174, 3274 N or Y D 11
3800 1, 3 Y 13, 14 3800 N D or S 12
3820 N 3820 N D
3900 Y 3900 N D, S, or S4
4100 Y 4100 N D, S, or S4
4245 Y 4245 N D
4248 Y 4248 N D
4250 Y 3174, 3274 N or Y D 11
7374 (by Y 5088 N D or S 12
5085)
7375 (by Y 5088 N D or S 12
5085)
Magnetic tape units
3420 3, 4, 5, 6, Y 15 3803 1, 2 Y D 15
7, 8
3422 Y 3422 Y D or S
3480 Y 3480 N D, S, or S4 25
3490 Y 39, 41, 46 3490 A0x, B0x, N D, S, or S4 25, 28, 38
D3x
3490E Y 40, 41, 46 3490E Ax0, Bx0, N D, S, or S4 25, 28, 38
Cxx, D4x
3590 Y 45, 46 3590 28, 43
40, 45, 46 3591 28, 43
Card readers and punches
2501 B1, B2 Y 2501 B1, B2 N D
2540 1 Y 16 2821 1, 5, 6 N D
3505 Y 3505 B1, B2 N D
3525 Y 3505 B1, B2 N D
Appendix D. List of I/O devices and control units 247

Data transmission multiplexers
uuuuu N 20 2701 1 N D
3704 N 18 3704 All N D
uuuuu N 21 3704 All N D
3705 N 18 3705 All N D
(3274A) 18
uuuuu N 21 3705 All N D
Communication controllers
3725 N 18 3725 N D
uuuu N 21 3725 N D
3745 N 18 3745 N D or S
29, 42 OSN
uuuu N 21 3745 N D or S
3746 N 44 3746 N D or S
uuuu N 44 3746 N D or S
3791L N 18 3791L N D
3791L, Y RS6K, 3791L, N D, S, or S4
3174 3174
(3174 SNA N 18 3791L N D or S
parallel
channel)
(3174 N 18, 37 3791L Y D or S
Token
Ring
Gateway)
(3174 N 18, 37 3791L Y D or S
ISDN
Gateway)
(3274A) N 18 3791L N D
Series/1*
uuuu Y 33 uuuu N D 33
Other
CTC Y 19, 22 CTC N D
3088 1, 2 N D, S, or S4 23
3172, Y 3172 N D, S, or S4 47
CTC, 3088
RS6K, Y RS6K, CTC, N D, S, or S4
CTC, 3088 3088
3995 Y 3995 N D, S, or S4 28, 32
SCTC Y 34 SCTC
BCTC Y 35 SCTC
FCTC Y 34 FCTC

3848 Y 3848 1 N D or S
CFS CFS
OSA 29, 42 OSA
OSAD OSA, OSN
OSN 29, 42 OSN
ESCON Directors
SWCH Y 26 9032 All 27, 28, 48
9033 27, 28, 48
SWCH 27, 28, 48
FICON Directors
SWCH Y 26 2032 27, 28
2042 27, 28
SWCH 27, 28
Table 15. z/OS considerations:

Several of the following notes apply to HCD only when migrating IOCP input (see Notes 16 on page 250, 17 on
page 250, 18 on page 250, 20 on page 251, 21 on page 251, 24 on page 251, 33 on page 251, 39 on page 251, 40 on
page 251, and 49 on page 252).
Notes:
1. IOCP generates only the specified number of devices. IOCP does not generate any extra devices
based on the device type. You must specify all the devices that exist for the given device type.
2. For the 3880, one CNTLUNIT statement must be specified for each Storage Director in the 3880. For
the 3990, one CNTLUNIT statement must be specified for each multipath storage director in the
3990.
3. Each channel path attaching to the 3880 Model 1, 2, 3, or 11 can operate independently in either
data-streaming mode or offset interlock mode. The 3880 has eight channel speed control switches for
setting the mode of operation. Regardless of the device types attached to the 3880: (a) set all 3880
channel speed control switches for channel paths attached to the CPC to data-streaming mode; (b)
specify PROTOCL=S in the CNTLUNIT statements for each 3880 Storage Director. These actions
allow the maximum data transfer rates and the use of longer cable lengths.
The CPC requires all control units that attach a common device to the CPC to use the same interface
protocol. Therefore, when both a 3880 and a 3830 attach a common device to the CPC, the user must
specify PROTOCL=D on both the CNTLUNIT statement for the 3830 storage director and the
CNTLUNIT statements for the attached 3880 Storage Directors. In this situation, you must set the
3880 channel speed control switches for the channel paths that attach the 3880 Storage Directors to
the non-data-streaming mode.
4. For the 3330V (virtual volumes for the MSS using 3350 devices as staging drives), you must also
specify the 3851 in an IODEVICE statement and an associated CNTLUNIT statement.
5. Specify UNIT=3350P for each 3350 that is attached to the paging storage director of a 3880 Model 11.
Specify UNIT=3350 for each 3350 that is attached to the non-paging storage director of a 3880 Model
11, to a 3880 Model 1 or 2, or to a 3830.
6. Specify UNITADD=((aa,32)). Bits 3-7 of the binary value of the byte representing aa must be zeros.

To identify the paging storage director of a 3880 Model 11 in the configuration reports, you can
specify UNIT=3880P. The non-paging storage director of a 3880 Model 11 can be specified
UNIT=3880.
7. Specify UNIT=3351P for each 3350 that is attached to the paging storage director of a 3880 Model 21.
8. Specify UNITADD=((aa,16)). Bits 4-7 of the binary value of the byte representing aa must be zeros.
To identify the paging storage director of a 3880 Model 21 in the configuration reports, you can
specify UNIT=3880P. The non-paging storage director of a 3880 Model 21 can be specified
UNIT=3880.
9. When you are using the 5081 in a 3250 compatibility mode, specify the 5081 as UNIT=2250,
MODEL=3. When you are using the 5081 in full function mode, specify the 5080 as UNIT=HFGD.
For more information, see Graphics Access Method/System Product: Initialization, Resource Definition, and
Customization, SC33-0141.
The 5088 device can run either in DC interlock protocol or in data-streaming protocol, depending on
the channel. Ensure that the interface protocol specified matches the protocol specified in the 5088.
10. For a 3272 attached to a block-multiplexer channel path, use SHARED=Y if multiple devices are
attached to the control unit. SHARED=N is required if a 3272 is attached to a byte multiplexer
channel path.
When multiple devices are attached to the control unit and SHARED=N is specified, the possibility
of additional processor use in the control program increases because of deferred condition code 1
I/O interruptions. If a single device is attached to the control unit, SHARED=N can be specified
without increasing the processor use in the control program for interruption processing.
11. For a 3174 or 3274 attached to a block-multiplexer channel path, use SHARED=Y if multiple devices
are attached to the control unit. If necessary, you can specify SHARED=N when multiple devices are
attached to the control unit. However, specifying SHARED=N might result in a control unit
malfunction or increase the possibility of additional processor use in the control program resulting
from deferred condition code 1 I/O interruptions.
If a single device is attached to the control unit (as is always the case for the 3174 (SNA) or 3274
Model 1A), SHARED=N can be specified and does not increase the processor use in the control
program for interruption processing.
12. The control unit can run in either DC interlock protocol or data-streaming protocol, depending on
the channel. Ensure that the interface protocol specified matches the protocol specified in the
CNTLUNIT statement.
13. Attach the 3800 to a block-multiplexer channel (TYPE=CVC) for best system performance.
14. Specify TIMEOUT=Y when IBM-supplied programs are used with the 3800. Non-IBM programs
might require TIMEOUT=N for the 3800 because I/O requests might require more than 4 seconds to
complete.
15. If the magnetic tape subsystem contains two to four control units that access the same group of
magnetic tape units, define each control unit separately. Each of the control units must have a
different channel path identifier, but all of the control units must specify the same address range.
Specify in the IODEVICE statement that the magnetic tape units are attached to all of the control
units. (The magnetic tape units might actually be only logically attached, but they are specified as
though they are physically attached.)
16. For HCD users, specify the 2540 with two IODEVICE statements: one with UNIT=2540R, the other
with UNIT=2540P. See Table 15 on page 249..
17. For HCD users, for the 3203, only the Model 5 is supported. See Table 15 on page 249.
18. For HCD users, a 3174 (SNA) or 3274 Model A is specified as a 3791L in the IODEVICE statement.
See Table 15 on page 249.
19. For the channel-to-channel adapter (CTCA), specify TIMEOUT=N because I/O sequences on the
CTCA might require more than 4 seconds to complete. Because TIMEOUT=N, the CTCA cannot be
attached to a byte multiplexer channel path.

20. For HCD users, in the UNIT parameter, you can specify any device or line that is attached to the
2701. See Table 15 on page 249.
21. For HCD users, in the UNIT parameter, you can specify any device or line that is attached to the
3704, 3705, 3725, and 3745. (This statement applies only to emulation program [EP] mode and byte
multiplexer mode.) See Table 15 on page 249.
22. For a 3088, specify UNITADD=b0 and ADDRESS=(aaa,n), where n=32 or n=64, depending on the
number of unit addresses required.
23. For a 3088, specify UNITADD=((a0,n)), where n=32 or n=64, depending on the number of unit
addresses required.
24. For HCD users, the UNIT type must indicate the mode of operation for the device. If the device is
set to 3380 track compatibility mode, use UNIT=3380. See Table 15 on page 249.
25. For a 3480 (all models) or a 3490 or 3490E (Models Axx or Bxx), specify UNITADD=((a0,16)).
26. For 2032, 2042, 9032, 9033, and SWCH, always code UNITADD=00.
27. For 2032, 2042, 9032, 9033, and SWCH, always code LINK=FE.
28. The 2105, 3490, 3490E, 3590, 3591, 3990, 3995, 9032, 9033, and SWCH control units require the unit
address range to begin with 00 for ESCON channel paths CNC and FCV. The 2032, 2042, 2105, and
3590 control units require the unit address range to begin with 00 for FICON channel path type FC.
The 3990 requires the unit address range to begin with X'10' or X'20' if DASD string 0 is not installed.
For more information, see 3990 Storage Control Planning, Installation, and Storage Administration Guide,
GA32-0100.
29. For devices assigned to OSC, OSD, OSE, OSM, OSN, or OSX channel paths, do not specify the
STADET keyword because it is ignored.
30. Specify one CNTLUNIT statement for each storage cluster in the IBM 9340 Direct Access Storage
Subsystem.
31. Specify UNITADD =((nn,32)). Bits 3-7 of the binary value of the byte representing nn must be zeros.
32. Specify UNITADD=((a0,8)) or, if a Model 111 or Model 112 expansion unit is attached, specify
UNITADD=((a0,16)), where a is any hexadecimal digit unless the channel paths are ESCON. For
ESCON channel paths, specify UNITADD=((00,8)) or UNITADD=((00,16)).
33. For HCD users, you can program a Series/1 to emulate a control unit with devices. In the
CNTLUNIT statement, specify in the UNIT keyword the type of control unit being emulated. In the
IODEVICE statement, specify in the UNIT keyword the type of device being emulated. See Table 15
on page 249.
34. If you specify UNIT=SCTC for devices assigned to ESCON CTC channels, the I/O subsystem uses
extended mode CTC. The I/O subsystem always uses extended mode CTC for devices assigned to
FICON CTC control units.
35. To run applications that require parallel basic mode CTC support using ESCON channels, specify
UNIT=BCTC. Only the ESCON CTC channel end of the ESCON CTC connection needs the
UNIT=BCTC specification. If UNIT=BCTC is not specified, the I/O subsystem uses extended mode
CTC.
36. The description is valid only for devices on channel-attached non-SNA 3174s and 3274s. Devices on
channel-attached SNA 3x74s are not defined separately so only the SNA control unit must be coded
in an IODEVICE statement. See the “Communication Controllers” section in Table 14 on page 245.
37. 3174s operating as a Token Ring or ISDN gateway must have only one IODEVICE and one
CNTLUNIT statement. The number value must be coded for both the ADDRESS= parameter in the
IODEVICE statement and the UNITADD= parameter in the CNTLUNIT statement to match the
address range the 3174 gateway supports.
38. For a 3490 or 3490E (Models Cxx or Dxx), specify UNITADD=((a0,2)).
39. For HCD users, specify UNIT=3480. See Table 15 on page 249.
40. For HCD users, specify UNIT=3490. See Table 15 on page 249.

41. For a 3490 or 3490E (Models Axx or Bxx) that is part of a 3495 Tape Library Dataserver, specify
UNITADD=b0 and ADDRESS=(aaa,16) regardless of the number of devices installed. For a 3490E
(Models Cxx) that is part of a 3494 Tape Library Dataserver, specify UNITADD=b0 and
ADDRESS=(aaa,2).
42. For information about configuring devices for OSA channels, see “OSA configuration considerations”
on page 33.
Note: OSA channels identify themselves as machine type 1730.

For further information about OSA configuration requirements for client server applications, see
Open Systems Adapter-Express Integrated Console Controller User's Guide, SA22-7935. For additional
information about OSA-ICC configuration requirements, see Open Systems Adapter-Express Customer's
Guide and Reference, SA22-7990.
43. For a 3590 or 3591, specify UNITADD=((00,4)). For a 3590 Model A60, specify UNITADD=((00,x)),
where x is the number of devices supported by the tape controller regardless of the number of
devices installed.
44. See the following documents for information about the 3746:
v 3745 Communication Controller Model A and 3746 Nways Multiprotocol Controller Models 900 and 950
Planning Series: ESCON Channels, GA27-4237
v 3745 Communication Controller Model A and 3746 Nways Multiprotocol Controller Models 900 and 950
Planning Series: Overview, Installation, and Integration, GA27-4234.
45. For a 3590 that is part of a 3494 or 3495 Tape Library Dataserver, specify UNITADD=00 and
ADDRESS=(aaa,4) regardless of the number of devices installed. For a 3590 Model A60, specify
UNITADD=00 and ADDRESS=(aaa,x) where x is the number of devices supported by the tape
controller regardless of the number of devices installed.
46. For a 3490 Model A02 and a 3490E Model A20, specify a preferred path for each device so that the
devices are divided between the two control unit functions. For example, you can specify the devices
with even unit addresses to have preferred paths to the first control unit and unit addresses with
odd unit addresses to have preferred paths to the second control unit. Also, if a control unit function
has multiple paths, assign the paths equally as preferred paths to the devices.
For a 3590, all 3490 models other than A02 and all 3490E models other than A20 do not specify a
preferred path.
47. Specify UNITADD=((nn,8)).
48. For maximum availability of a control unit attached to an ESCON Director (including the control
unit associated with LINK address FE), consider these guidelines in the following order of priority:
v Attach each channel path from the same CPC to separate ESCON Directors
v Attach each channel path to a separate port card quadrant in the ESCON Director
v Attach each channel path within the same quadrant to a separate port card in the ESCON
Director.
49. For HCD users with 3380 format devices attached to a 2105, you can specify 3380 parallel access
volume (PAV) base devices as 3380B and aliases as 3380A. For 3390 format devices attached to a
2105, you can specify 3390 PAV base devices as 3390B and aliases as 3390A.
50. For the 2105, one CNTLUNIT statement must be specified for each logical subsystem (LSS). Specify
the CUADD keyword with a value that matches the CU image address associated with the LSS.

Appendix E. Configuration examples
This appendix presents several examples to illustrate key concepts associated with the different types of
channel paths that IOCP supports. The IOCP inputs included with each figure show the correct way to
code the configuration. The examples do not always illustrate configurations designed for reliability,
availability, or serviceability.
More than eight paths to an ESCON device
Figure 35. LPAR mode with more than eight paths to an ESCON device
IOCP input for the configuration

RESOURCE PART=(LP1,LP2,LP3)
CHPID PCHID=100,PATH=22,TYPE=CNC,PART=LP1,SWITCH=03
CHPID PCHID=109,PATH=33 TYPE=CNC,PART=LP3,SWITCH=03
CHPID PCHID=10A,PATH=34,TYPE=CNC,PART=LP3,SWITCH=03
CHPID PCHID=10B,PATH=35,TYPE=CNC,PART=LP3,SWITCH=03
CNTLUNIT CUNUMBR=000,PATH=(22,23,24,25),UNITADD=((00,16)),UNIT=XXXX, X
LINK=(C5,C5,C5,C5)
LINK=(C5,C5,C5,C5)

LINK=(C5,C5,C5,C5)
IODEVICE ADDRESS=(100,16),CUNUMBR=000,UNIT=XXXX
Shared paths to an ESCON device
LP1 LP2 LP3
LPAR with MIF
22 24 28 32
Director
(SWITCH=03)
C5
Control
Unit
UA 00-0F
Devices 100-10F
Figure 36. LPAR mode with shared paths to an ESCON device

RESOURCE PARTITION=(LP1,LP2,LP3)
CHPID PCHID=100,PATH=22,TYPE=CNC,SWITCH=03,SHARED
LINK=(C5,C5,C5,C5)
ESCON CTC point-to-point

Figure 37. ESCON CTC point-to-point
IOCP input for system 1 configuration

RESOURCE PART=SYSTEM1
CHPID PCHID=100,PATH=(31),TYPE=CNC,PART=SYSTEM1
CNTLUNIT CUNUMBR=000,PATH=(31),UNITADD=((00,1)),UNIT=SCTC
IODEVICE ADDRESS=100,CUNUMBR=000,UNIT=SCTC

CHPID PCHID=100,PATH=(22),TYPE=CNC,PART=SYSTEM2
CNTLUNIT CUNUMBR=100,PATH=(22),UNITADD=((00,1)),UNIT=SCTC
Note: You must specify the same unit address for devices from different logical partitions to
communicate with each other (00 in this example). This example assumes that the application program
supports send and receive functions on a single device number.
ESCON CTC dynamic through an ESCON director with multiple paths

Figure 38. ESCON CTC dynamic through an ESCON director with multiple paths

RESOURCE PART=(LPAR1,LPAR2,LPAR3)
CHPID PATH=(22),TYPE=CTC,SWITCH=01,PART=LPAR1,PCHID=100
CHPID PATH=(33),TYPE=CNC,SWITCH=01,PART=LPAR2,PCHID=101
CHPID PATH=(39),TYPE=CNC,SWITCH=01,PART=LPAR3,PCHID=102
CNTLUNIT CUNUMBR=000,PATH=(22),UNITADD=((00,1)),UNIT=SCTC, X
LINK=D0
LINK=E0
LINK=C0
LINK=C0
communicate with each other. In this example, device 100 in LPAR1 can communicate with device 200 in
LPAR2. Also, device 110 in LPAR1 can communicate with device 210 in LPAR3. This example assumes
that the application program supports send and receive functions on a single device number.
FICON CTC and director with single path

System
LP1 LP2 LP3

CHP33
D0
Director
(SWITCH=01)
Figure 39. FICON CTC and director with single path

RESOURCE PARTITION=((LP1,1),(LP2,2),(LP3,3))
*
CHPID PATH=(33),TYPE=FC,SWITCH=01,PART=(LP1,LP2,LP3),SHARED,PCHID=180
*
CNTLUNIT CUNUMBR=4010,PATH=33,LINK=D0,UNIT=FCTC,UNITADD=((00,2)),CUADD=1
IODEVICE ADDRESS=(4010,2),CUNUMBR=4010,UNIT=FCTC,UNITADD=00,NOTPART=LP1
*
*
communicate with each other. In this example, each logical partition can communicate with the other
logical partitions. This example defines separate send and receive devices to be used by an application
program as follows:
v LP1 sends data to LP2 with device 4020 and LP3 with device 4030. LP2 and LP3 both receive data from
LP1 with device 4010.
v LP2 and LP3 send data to LP1 with device 4011. LP1 receives data from LP2 with device 4021 and
from LP3 with device 4031.
v LP2 sends data to LP3 with device 4030. LP3 receives data from LP2 with device 4020.
v LP3 sends data to LP2 with device 4021. LP2 receives data from LP3 with device 4031.
Chained ESCON directors

Figure 40. Chained ESCON directors

CHPID PCHID=100,PATH=(31),TYPE=CNC,SWITCH=01,PART=SYSTEM1
CNTLUNIT CUNUMBR=002,PATH=(31),UNITADD=((00,1)),UNIT=XXX, X
LINK=E0
LINK=F0
Note: Channel path (chpid) 31 has a dedicated connection through ESCON Director 00 and two dynamic
connections through ESCON Director 01.

CHPID PCHID=180,PATH=(22),TYPE=CNC,SWITCH=00,PART=SYSTEM2
LINK=D0
LINK=C0
Note: Channel path (chpid) 22 has two dynamic connections through ESCON Director 00 and has no
connection to ESCON Director 01.
Cascaded FICON directors

System
CHP 30 CHP 31 CHP 32
Director Director
(Switch Address=21) (Switch Address=22)
C0 C0
CU 001 CU 002
Figure 41. Cascaded FICON directors

RESOURCE PART=(LP1,LP2,LP3)
CHPID PCHID=100,PATH=30,TYPE=FC,SWITCH=21,SHARED
CNTLUNIT CUNUMBR=001,PATH=(30,31,32),LINK=(C0,21C0,21C0),UNIT=XXX,...
CNTLUNIT CUNUMBR=002,PATH=(31,32),LINK=(22C0,22C0),UNIT=XXX,...
Note: In this example, the switch numbers specified for the FICON Directors match the switch addresses
assigned to the Directors through their switch control panels.
In this example, chpids 30, 31, and 32 are attached to control unit 001. Chpid 30 communicates only to
control units attached to switch 21 so it has used a 2-digit link address (C0) even though the switch is
configured with the high integrity attributes and supports 4-digit link addresses. Chpids 31 and 32
communicate to control units that are not connected to their entry switches so they must have 4-digit link
addresses. The first two digits identify the switch address and the last two digits identify the port
address to which the control unit is attached. Link address 21C0 identifies port C0 on the switch with
switch address 21. Link address 22C0 identifies port C0 on the switch with switch address 22. I/O for
control unit 001 for chpid 32 flows through both switches while chpid 31 uses only switch 21. I/O for
control unit 002 for chpid 31 flows through both switches while chpid 32 uses only switch 22.
Note: This example does not follow all recommendations. For example, chpid 31 connects to switch 21
and uses 4-digit link addresses while chpid 30 also connects to switch 21 but uses 2-digit link addresses.
When a 4-digit link address is specified for a channel path, ensure that all channel paths connected to the
same entry switch specify 4-digit link addresses so IOCP can perform better checking of the switch
configuration. See “Defining FICON switch configurations” on page 38 for more information about
cascading switches and specifying link addresses for FC channel paths.
Control unit logical addressing

Figure 42. Control unit logical addressing

CHPID PATH=(21),TYPE=CNC,SWITCH=02,...
CNTLUNIT CUNUMBR=000,PATH=(21,22),UNITADD=((00,16)),UNIT=XXXX, X
LINK=(F0,F0),CUADD=0
LINK=(E0,E0),CUADD=0
Channels through an ESCON converter

System
CHP20 CHP21
Fiber
Optic
Cable
Director
(Switch=00)
F0
ESCON ESCON
Converter Converter
Bus-and-Tag
Cable
CU 000 UA 00-0F
CU 001 UA 20-2F
Figure 43. Channels through an ESCON converter

RESOURCE PART=LP1
CHPID PCHID=100,PATH=20,TYPE=CVC,PART=LP1
CHPID PCHID=101,PATH=21,TYPE=CVC,PART=LP1,SWITCH=00
CNTLUNIT CUNUMBR=000,PATH=(20,21),UNITADD=((00,16)),UNIT=XXX, X
LINK=(**,F0),SHARED=N
CNTLUNIT CUNUMBR=001,PATH=(20,21),UNITADD=((20,16)),UNIT=XXX, X
LINK=(**,F0),SHARED=N
FCV channel configuration

System
CHP00 CHP20 CHPF4
ESCON FICON
I/O I/O
Interface Interface
ESCON Director
(Switch=10)
F0
ESCON
I/O
Interface
CU 001 UA 00-0F
Figure 44. FCV channel configuration

CHPID PATH=(00),TYPE=CNC,...
CHPID PATH=(F4),TYPE=FCV,SWITCH=10,...
CNTLUNIT CUNUMBR=001,PATH=(00,20,F4),LINK=(**,F0,F0),UNIT=XXX, X
UNITADD=((00,16))
ESCON control units requiring unit address 00
Figure 45. ESCON control units requiring unit address 00
OSA configuration with two logical partitions

Two different home IP addresses are specified in the TCP/IP profile and two different TRLE (transport
resource list element) statements are defined. Because the channel path is shared, both logical partitions
have access to the READ, WRITE, and DATAPATH devices.

IOCP Input
Figure 46. OSA configuration with two logical partitions

RESOURCE PART=(LPLEFT,LPRIGHT)
CHPID PATH=F8,TYPE=OSD,PART=((LPLEFT,LPRIGHT),(=)),PCHID=180
CNTLUNIT CUNUMBR=F800,PATH=F8,UNIT=OSA
IODEVICE ADDRESS=(590,3),CUNUMBR=F800,UNIT=OSA
Additional OSA configuration examples

For additional OSA configuration examples, see Open Systems Adapter-Express Customer's Guide and
Reference, SA22-7935.
OSA-ICC configuration

System
CSS 0 CSS 1
LP01 LP02 MVS1
CHP 80
Network
3270 Client
Figure 47. OSA-ICC

RESOURCE PART=((CSS(0),(LP01,1),(LP02,2)),(CSS(1),(MVS1,2)))
CHPID PCHID=1C0,PATH=(CSS(0,1),80),TYPE=OSC
CNTLUNIT CUNUMBR=1000,PATH=((CSS(0),80),(CSS(1),80)),UNIT=OSC
IODEVICE ADDRESS=(2400,80),CUNUMBR=1000,UNIT=3270,MODEL=X
Partial manual session configuration file

<CONFIG_SESSION>
<SESSION1>
CLIENT_IP= 10.10.10.15
CSS= 00 MIFID= 01 DEVICE= 2400
GROUP= "OSA01"
CONSOLE_TYPE= 2 RESPONSE= ON READ_TIMEOUT= 10
</SESSION1>
<SESSION2>
CLIENT_IP= 10.10.10.15
GROUP= "OSA01"
</SESSION2>
<SESSION3>
CLIENT_IP= 10.10.10.15
GROUP= "OSA02"
</SESSION3>
.
.
.
<SESSION9>
CLIENT_IP= 10.10.10.17
GROUP= "OSA01"
</SESSION9>

<SESSION10>
CLIENT_IP= 10.10.10.18
GROUP= "OSAXX13" CONSOLE_TYPE= 1 RESPONSE= ON READ_TIMEOUT= 10
</SESSION10>
</CONFIG_SESSION>
Additional OSA-ICC configuration examples

For additional OSA-ICC configuration examples, see Open Systems Adapter-Express Integrated Console
Controller User's Guide, SA22-7990.
FCP channel configuration
CPC
Host Operating System
Host Application
SCSI Stack
QDIO Driver
QDIO
FCP Channel
Adapter
Switch
Destination N_Port Destination N_Port

(345adf...) (500507...)
Logical Logical Logical

Unit Unit Unit
4209... 420b... 562b...
Figure 48. FCP channel configuration

Partial host operating system configuration definition
0x 1200 ...:0x345adf3322443525 ...:0x4209000000000000
0x 1200 ...:0x345adf3322443525 ...:0x420b000000000000
0x 1200 ...:0x5005076300c40c85 ...:0x562b000000000000

RESOURCE...
CHPID PATH=80,TYPE=FCP,...
CNTLUNIT CUNUMBR=1111,PATH=80,UNIT=...
IODEVICE ADDRESS=1200,CUNUMBR=1111,UNIT=...
Overview of the configuration definitions

A host operating system definition consists of multiple fields. The preceding example attempts to show
the primary fields that are required to define FCP devices in a host operating system. It specifies two
logical units for one target device and one logical unit for another target device. The first field shown is
the device number (for example, 0x1200), which is the same device number specified in the IOCP
definition. The next field shown is the worldwide port number, identifying the Fibre Channel port of the
SCSI target device (for example, 0x345adf...). The final field shown is the logical unit number (for
example, 0x4209... and 0x420b...).
The IOCP definition consists of the FCP channel path, a control unit, which is only necessary so that
devices can be defined, and a device. The device 1200 is associated with a subchannel in the logical
channel subsystem (CSS).
A host application issues an I/O request. The SCSI/FCP stack gets the CSS subchannel number for the
device number (1200) and packages the FCP I/O operation into a format acceptable for QDIO operations
and passes it to the QDIO driver. The QDIO driver uses queues associated with the subchannel for device
1200 and performs the QDIO operation to the FCP channel adapter. The FCP channel performs the Fibre
Channel Protocol operation with the correct destination logical unit (LUN) using the correct Fibre
Channel port of the SCSI target device.
Coupling facility configuration
Figure 49. Coupling facility configuration

IOCP input for z/OS configuration
RESOURCE PART=(MVS1,MVS2)
CHPID PATH=70,TYPE=CFS,PART=(MVS1,MVS2),PCHID=210
CHPID PATH=73,TYPE=CFS,PART=(MVS1,MVS2),PCHID=211
CNTLUNIT CUNUMBR=FFE,PATH=(70,73),UNIT=CFS
IODEVICE ADDRESS=(FFF8,2),CUNUMBR=FFE,UNIT=CFS
IODEVICE ADDRESS=(FFFC,2),CUNUMBR=FFE,UNIT=CFS
This example also applies to channel path types CBS, CFP, and CBP. A corresponding change would be
required in the coupling facility configuration.
IOCP input for coupling facility configuration

RESOURCE PART=CF1
CHPID PATH=16,TYPE=CFR,PART=CF1,PCHID=300
CHPID PATH=17,TYPE=CFR,PART=CF1,PCHID=301
ESCON CTC with shared and unshared channels

CNC CHPID 50 and CTC CHPID 51 in System A are shared by LP1A, LP2A, LP3A, and LP4A. All
partitions in the figure participate in an any-to-any CTC configuration. No redundancy is provided. All
the channel paths on System B are unshared. For recommendations for numbering CTC control units and
devices to simplify CTC configurations, see the ESCON and FICON Channel-to-Channel Reference,
SB10-7034.
Four CTC communications are defined between each LP except for LP4A, which has six connections to
each of the other LPs. See Figure 50 on page 268.
Note: All shared CTC examples within a single CPC exclude an LP from communicating with itself. This
exclusion is required for some subsystems (for example, XCF). Other subsystems might not have this
restriction and might support the communication of an LP with itself.

Figure 50. ESCON CTC with shared and unshared channels
IOCP input for System A (shared)

RESOURCE PARTITION=((LP1A,1),(LP2A,2),(LP3A,3),(LP4A,4))
CHPID PATH=50,TYPE=CNC,SWITCH=01,SHARED,...
CHPID PATH=51,TYPE=CTC,SWITCH=01,SHARED,...
*********************************************************************
* CTC wrap-around connections among LP1A, LP2A, LP3A, and LP4A
* Via CHPID 50 (CNC) in System A to CHPID 51 (CTC) in System A
*
* Define a control unit representing each partition sharing
* destination channel CHPID 51 (CTC) to source channel CHPID 50 (CNC).
*
* Representing LP1A:
CNTLUNIT CUNUMBR=5010,PATH=50,LINK=C1,CUADD=1,UNITADD=((40,6)),UNIT=SCTC
* LP1A is excluded from PART keyword to disallow LP1A to LP1A CTC
IODEVICE ADDRESS=(5010,4),CUNUMBR=5010,UNITADD=40,PART=(LP2A,LP3A,LP4A),UNIT=SCTC
IODEVICE ADDRESS=(5014,2),CUNUMBR=5010,UNITADD=44,PART=LP4A,UNIT=SCTC
*
IODEVICE ADDRESS=(5024,2),CUNUMBR=5020,UNITADD=44,PARTITION=LP4A,UNIT=SCTC
*

*
*
*
* destination channel CHPID 50 (CNC) to source channel CHPID 51 (CTC).
*
*
*
*****************************************************************
* CTC connections between (LP1A,LP2A,LP3A,LP4A) and LP5B
*
* Define a control unit representing destination channel CHPID 60 (CNC)
* to source channel CHPID 51 (CTC).
CNTLUNIT CUNUMBR=5150,PATH=51,LINK=D0,CUADD=0,UNITADD=((40,6)),UNIT=SCTC
IODEVICE ADDRESS=(5150,4),CUNUMBR=5150,UNITADD=40,UNIT=SCTC
*****************************************************************
*
*****************************************************************
*
* Define a control unit representing destination channel CHPID 63 (CTC)
* to source channel CHPID 50 (CNC).
IOCP input for System B (unshared)

CHPID PATH=60,TYPE=CNC,SWITCH=01,...
CHPID PATH=61,TYPE=CNC,SWITCH=01,...
CHPID PATH=62,TYPE=CTC,SWITCH=01,...
CHPID PATH=63,TYPE=CTC,SWITCH=01,...
**************************************************************
* CTC wrap-around connections among LP5B, LP6B, and LP7B
*
CNTLUNIT CUNUMBR=6060,PATH=60,LINK=D2,UNITADD=((40,4)),UNIT=SCTC

*
*
*
*****************************************************************
* CTC connections between LP5B and (LP1A,LP2A,LP3A,LP4A)
* Via CHPID 60 (CNC) in System B to CHPID 51 (CTC) in System A
*
*
*****************************************************************
*
*
*****************************************************************
* Via CHPID 63 (CTC) in System B to CHPID 50 (CNC) in System A
*

*
ESCON CTC with shared channels

A fully redundant any-to-any CTC configuration consists of two sets of CTC minimum configurations,
each associated with a separate ESCON Director. CTC communication workload between System A and
System B is balanced between four CTC connections. All the channel paths are shared. See Figure 51 on
page 272. For recommendations for numbering CTC control units and devices to simplify CTC
configurations, see the ESCON and FICON Channel-to-Channel Reference, SB10-7034.
restriction and might support the communication of an LP with itself.

Figure 51. ESCON CTC with shared channels
IOCP input for System A (shared)

RESOURCE PARTITION=((LP1A,1),(LP2A,2),(LP3A,3),(LP4A,4))
***************************************************
* First any-to-any CTC connections with Switch=01 *
***************************************************
*
*********************************************************************

*
*
* LP1A is excluded using NOTPART keyword to disallow LP1A to LP1A CTC
IODEVICE ADDRESS=(5010,6),CUNUMBR=5010,UNITADD=40,NOTPART=(LP1A),UNIT=SCTC
*
*
*
*
*
*
*
************************************************************************
* CTC connections between (LP1A,LP2A) and (LP5B,LP6B,LP7B)
*
* Only LP1A and LP2A use 50-61 CTC connection for workload balance
* Representing LP5B:
IODEVICE ADDRESS=(5050,6),CUNUMBR=5050,UNITADD=40,PART=(LP1A,LP2A),UNIT=SCTC
*
*
*************************************************************************
*
* Only LP3A and LP4A use 51-60 CTC connection for workload balance

*
*
****************************************************
* Second any-to-any CTC connections with Switch=02 *
****************************************************
*
*********************************************************************
*
*
*
*
*
*
*
*
*
************************************************************************
*
* Only LP1A and LP2A use 52-63 CTC connections for workload balance

*
*
*************************************************************************
*
* Only LP3A and LP4A use 53-62 CTC connections for workload balance
IOCP input for System B (unshared)

RESOURCE PARTITION=((LP5B,5),(LP6B,6),(LP7B,7))
***************************************************
* First any-to-any CTC connections with Switch=01 *
***************************************************
*
**************************************************************
* CTC wrap-around connections among LP5B, LP6B, LP7B
* Via CHPID 60 (CNC) in System B to CHPID 61 (CTC) in System B
*
*
* LP5B is excluded using NOTPART keyword to disallow LP5B to LP5B CTC
IODEVICE ADDRESS=(6050,6),CUNUMBR=6050,UNITADD=40,NOTPART=(LP5B),UNIT=SCTC
*
*
*
*
*

*
**********************************************************************
* CTC connections between (LP5B,LP6B,LP7B) and (LP1A,LP2A)
*
*
CNTLUNIT CUNUMBR=6120,PATH=61,LINK=C0,CUADD=2,UNITADD=((40,6),UNIT=SCTC
**********************************************************************
*
*
****************************************************
* Second any-to-any CTC connections with Switch=02 *
****************************************************
*
**************************************************************
* CTC wrap-around connections among LP5B, LP6B, LP7B
* Via CHPID 62 (CNC) in System B to CHPID 63 (CTC) in System B
*
*
*
*
*

*
*
**********************************************************************
*
*
**********************************************************************
*
*
FICON CTC with shared channels

A fully redundant any-to-any CTC configuration consists of two sets of CTC minimum configurations,
each associated with a separate FICON Director. CTC communication is configured such that the
definitions operate in half duplex on the FICON channels. See Figure 52 on page 278. For
recommendations for numbering CTC control units and devices to simplify CTC configurations, see the
ESCON and FICON Channel-to-Channel Reference, SB10-7034.
restriction and might support an LP's communicating with itself.
This example uses a simple CTC configuration approach where the same channel path is always used for
sending or receiving data and the same destination link address for a control unit number is always used.
For example, System A always uses CHPID 50 on switch 01 for sending data and always uses CHPID 51
on switch 01 for receiving data. Also, all CPCs always use FCTC control unit 4010 to send data to LP1A
on System A through link address C1 (CHPID 51) on System A. So a single destination link address (C1)
is always associated with the same control unit number (4010).
This approach results in the FICON channels having half duplex definitions. That is, all FCTC I/O data
flow is on one fiber of the FICON channel resulting in its being half duplex. To balance I/O data flow on
the FICON channel in this case, put tape devices on the channels used for receive FCTC control units and
put DASD devices on the channels used for send FCTC control units. For example, you would put tape

devices on CHPID 51 in System A and DASD devices on CHPID 50 in System A.
System A System B
CHPID CHPID
Number FICON Director Number
LP1A Port Number LP1B
FC FC
Shared by 50 60 Shared by
1, 2, 3, 4 C0 D0 1, 2, 3
LP2A Switch=01 LP2B
FC C1 D1 FC
1, 2, 3, 4 1, 2, 3
LP3A LP3B
FC FC
1, 2, 3, 4 C2 D2 1, 2, 3
LP4A
Switch=02
FC C3 D3 FC
1, 2, 3, 4 1, 2, 3
Figure 52. FICON CTC (fully redundant)
Every logical partition has eight primary and eight secondary CTC connections with every other logical
partition. The primary connections use switch 01 and the secondary connections use switch 02.
IOCP statements for System A

RESOURCE PART=((LP1A,1),(LP2A,2),(LP3A,3),(LP4A,4))
*
CHPID PATH=50,TYPE=FC,SWITCH=01,SHARED,...
*
*****************************************************
* Primary any-to-any CTC connections with Switch=01 *
*****************************************************
*
**********************************************************************
* Primary wrap-around connections for logical partitions on System A *
**********************************************************************
*
* Send FCTC Control Unit and Device Definitions to CTC image ID 01
*
CNTLUNIT CUNUMBR=4010,PATH=50,LINK=C1,CUADD=1,UNITADD=((00,8)), +
UNIT=FCTC
IODEVICE ADDRESS=(4010,8),CUNUMBR=4010,UNIT=FCTC,UNITADD=00, +
NOTPART=LP1A
*
*
UNIT=FCTC
NOTPART=LP2A
*
*
UNIT=FCTC

NOTPART=LP3A
*
*
UNIT=FCTC
NOTPART=LP4A
*
* Receive FCTC CU and Device Definitions from CTC image ID 01
*
UNIT=FCTC
NOTPART=LP1A
*
*
UNIT=FCTC
NOTPART=LP2A
*
*
UNIT=FCTC
NOTPART=LP3A
*
*
UNIT=FCTC
NOTPART=LP4A
**********************************************************************
* Primary CTC connections on Switch=01 to System B. *
**********************************************************************
*
*
CNTLUNIT CUNUMBR=4210,PATH=50,LINK=D1,CUADD=1,UNITADD=((00,8)), +
UNIT=FCTC
IODEVICE ADDRESS=(4210,8),CUNUMBR=4210,UNIT=FCTC,UNITADD=00
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*

UNIT=FCTC
*
*
UNIT=FCTC
*
*******************************************************
* Secondary any-to-any CTC connections with Switch=02 *
*******************************************************
*
*********************************************************
* Secondary wrap-around connections for LPs on System A *
*********************************************************
*
*
UNIT=FCTC
NOTPART=LP1A
*
*
UNIT=FCTC
NOTPART=LP2A
*
*
UNIT=FCTC
NOTPART=LP3A
*
*
UNIT=FCTC
NOTPART=LP4A
*
*
UNIT=FCTC
NOTPART=LP1A
*
*
UNIT=FCTC
NOTPART=LP2A
*
*
UNIT=FCTC
NOTPART=LP3A
*

*
UNIT=FCTC
NOTPART=LP4A
***********************************************************************
* Secondary CTC connections on Switch=02 to System B. *
***********************************************************************
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
IOCP statements for System B

RESOURCE PART=((LP1B,1),(LP2B,2),(LP3B,3))
*
*
*****************************************************
* Primary any-to-any CTC connections with Switch=01 *
*****************************************************
*
**********************************************************************
* Primary wrap-around connections for logical partitions on System B *
**********************************************************************
*
*
UNIT=FCTC

NOTPART=LP1B
*
*
UNIT=FCTC
NOTPART=LP2B
*
*
UNIT=FCTC
NOTPART=LP3B
*
*
UNIT=FCTC
NOTPART=LP1B
*
*
UNIT=FCTC
NOTPART=LP2B
*
*
UNIT=FCTC
NOTPART=LP3B
**********************************************************************
* Primary CTC connections on Switch=01 to System A. *
**********************************************************************
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*

UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*******************************************************
* Secondary any-to-any CTC connections with Switch=02 *
*******************************************************
*
*********************************************************
* Secondary wrap-around connections for LPs on System B *
*********************************************************
*
*
UNIT=FCTC
NOTPART=LP1B
*
*
UNIT=FCTC
NOTPART=LP2B
*
*
UNIT=FCTC
NOTPART=LP3B
*
*
UNIT=FCTC
NOTPART=LP1B
*
*
UNIT=FCTC
NOTPART=LP2B
*
*

UNIT=FCTC
NOTPART=LP3B
***********************************************************************
* Secondary CTC connections on Switch=02 to System A. *
***********************************************************************
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC
*
*
UNIT=FCTC

Appendix F. Machine limits and rules
Machine rules
Table 16 lists the machine rules for the CPCs that IOCP supports.
| Table 16. Machine rules
| Machine type
| (See Note 1 on page 292)
| Machine rules 2084,1 2086,1 2094,1 2096,1 2097,1 2098,1 2817,1 2818,1 2827,1 2828,1 2964,1 2965,1 3906,1 3907,1
| Write IOCDS in Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
| preparation for
| CPC upgrade
| (CHECKCPC=NO
| or NOCHKCPC)
| CPC has fixed HSA No No No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
| Default system if No No No No No No No No No No No No No Yes
| IOCP is running on
| an unsupported
| processor
| FC and FCP No No No No No No No No No No No No Yes Yes
| PCHIDs on same
| adapter require
| MIXTYPE keyword
|
Note: See “MIXTYPE=” on page 70 for more information about the MIXTYPE keyword on the CHPID
statement.
To write an IOCDS in preparation for a CPC upgrade, you specify keyword CHECKCPC=NO in the
PARM parameter of the EXEC statement (z/OS or VSE) or CMS option NOCHKCPC. To prepare for an
upgrade, you can write an IOCDS to any CPC that requires the ICP version of IOCP. See Table 18 on
page 287 to determine which CPCs require an IOCDS written by ICP IOCP. You can also write an IOCDS
when preparing to upgrade from one of the following CPCs to a CPC that requires the ICP version of
IOCP:
v 2064 zSeries 900 model CPCs at EC J10638 or later
v 2066 zSeries 800 model CPCs.
The IOCDS cannot be used to power-on reset the z800 or z900 CPC until it is upgraded to the type of
CPC the IOCDS supports.
Note: IOCP can write an IOCDS in preparation for an upgrade to a CPC for which IBM does not offer an
upgrade. For example, you can write a SYSTEM=(2086,1) IOCDS to a 2084 CPC but no upgrade is offered
from a 2084 to a 2086 and the IOCDS is unusable on the 2084.

Machine limits
ICP IOCP supports the following limits:
Table 17. Constant machine limits
Machine limits
Control unit headers (CUHs)
Maximum channel paths per CUH 8
Physical control units (PCUs)
Maximum PCUs per CVC or CBY channel path (Note 1 on page 287) 48
Maximum PCUs per CFS, CFP, CBS, CBP, ICP, CIB, CL5, CS5, FCP, OSC, OSE, or OSN channel path 1
Maximum PCUs or link addresses per CNC or CTC channel path 120
Maximum PCUs or link addresses per FCV or FC channel path (Note 5 on page 287) 256
Maximum PCUs per IQD channel path 64
Maximum PCUs per OSD, OSM, or OSX channel path (Note 7 on page 287) 16
Maximum UA ranges per CVC or CBY channel path 64
Maximum UA ranges per shared CTC channel path 120
Maximum UA ranges per FC channel path (Note 10 on page 293) 256
Maximum UAs per CNC channel path 1024
Maximum UAs per CTC channel path 512
Maximum UAs per FCV channel path 16 K*
Maximum channel paths per PCU (Note 3 on page 287) 8
Devices
Maximum devices per CVC, CBY, CFS, CFP, CBS, CBP, CIB, CL5, CS5, or ICP channel path 256
Maximum devices per CNC channel path 1024
Maximum devices per CTC channel path 512
Maximum devices per OSC channel path (Note 6 on page 287) 254
Maximum devices per OSE channel path (Note 6 on page 287) 255
Maximum devices or subchannels per OSD or OSX channel path with priority specification enabled 480
Maximum devices or subchannels per OSD, OSM, or OSX channel path with priority specification 1920
disabled (CHPARM=02; Note 8 on page 287)
Maximum UNIT=OSN devices or subchannels per OSN channel path 480
Maximum UNIT=3745 devices or subchannels per OSN channel path 180
Maximum devices or subchannels per FCP channel path 480
Maximum devices or subchannels per FCV channel path 16 K
Maximum devices or subchannels for all IQD channel paths 12 K
Maximum channel paths per device (Note 4 on page 287) 8
Maximum PCUs per device (Note 4 on page 287) 8
Channel paths
Valid chpid range per CSS (Note 2 on page 287) 00-FF
Maximum channel paths per CIB AID 16
Maximum channel paths per CS5 AID port 4
Maximum channel paths per CL5 PCHID port 4

Table 17. Constant machine limits (continued)
Machine limits
Maximum port value for CL5 or CS5 2
IOCDS Levels A0-A3
Functions
Maximum port value for HYL or ROC2 2
Logical partitions
Maximum logical partitions per CSS (Note 9) 15
*K=1024
Notes:
1. The IOCP maximum of 48 allows you to over-define control units (possibly on a switch) for a path.
However, do not use more than eight control units on a path at one time.
2. All chpid range values are hexadecimal. You can specify the maximum chpid range even though your
CPC might have only a subset of the channel paths installed.
3. If you define the PCU as an ESCON CTC (UNIT=SCTC) or FICON CTC (UNIT=FCTC) control unit or
attach the PCU to an FCP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel path, the PCU can
attach to only one channel path per CSS.
4. If the device attaches to a PCU that you defined as an ESCON CTC or FICON CTC control unit or the
PCU attaches to an FCP, IQD, OSC, OSD, OSE, OSM, OSN, or OSX channel path, the device can
attach to only one PCU and to one channel path per CSS.
If the device attaches to a PCU attached to CFS, CFP, CBS, CBP, ICP, CIB, CL5, CS5 channel paths, the
device can attach to only one PCU.
5. FCV channel paths can have link addresses in the range 01-FE and, therefore, have a maximum of 254
link addresses.
6. You can use a maximum of 240 OSA valid subchannels on an OSE channel path even though IOCP
permits you to define more than this number. You can use a maximum of 120 valid subchannels on an
OSC channel path, even though IOCP permits you to define more than this number. See “OSA
configuration considerations” on page 33 for more information.
7. Multiple control units are supported on OSD channel paths on 2084 and 2086 models at EC J13486
(with an LMC patch installed). Otherwise, a maximum of one physical control unit can be defined on
an OSD channel path. Multiple control units are allowed on OSM and OSX channel paths on 2817 and
later models.
8. Disabling priority specification (for example, specifying CHPARM=02 on the CHPID statement) for an
OSD channel path is supported on these channel paths on 2084 and 2086 models at EC J13486 (with
an LMC patch installed). OSA-Express2 and later features are required when disabling priority
specification on OSD. OSM channel paths always operate with priority queuing disabled. OSX and
OSM channel paths require OSA-Express3 or later features.
| 9. On 3907, 3906, 2965 and 2964 models, the maximum CSS ID supported by the CPC can only contain
| 10 logical partitions with MIF IDs in the range 1-A.
ICP IOCP supports the following limits for the following CPCs (machine types):
| Table 18. Machine limits
| Machine type (Note 1 on page 292)
| Machine Limit 2084,1 2086,1 2094,1 2096,1 2097,1 2098,1 2817,1 2818,1 2827,1 2828,1 2964,1 2965,1 3906,1 3907,1
| Devices or subchannels
Appendix F. Machine limits and rules 287

| Table 18. Machine limits (continued)
| Machine Limit 2084,1 2086,1 2094,1 2096,1 2097,1 2098,1 2817,1 2818,1 2827,1 2828,1 2964,1 2965,1 3906,1 3907,1
| Maximum 64512 64512 65280 65280 65280 65280 65280 65280 65280 65280 65280 65280 65280 65280
| devices or (63 (63 (63.75 (63.75 (63.75 (63.75 (63.75 (63.75 (63.75 (63.75 (63.75 (63.75 (63.75 (63.75
| subchannels per K*) K*) K*) K*) K*) K*) K*) K*) K*) K*) K*) K*) K*) K*)
| CSS for
| subchannel set 0
| Maximum N/A N/A 65535 65535 65535 65535 65535 65535 65535 65535 65535 65535 65535 65535
| devices or (64 (64 (64 (64 (64 (64 (64 (64 (64 (64 (64 (64
| subchannels per K*-1) K*-1) K*-1) K*-1) K*-1) K*-1) K*-1) K*-1) K*-1) K*-1) K*-1) K*-1)
| CSS for
| subchannel set
| other than 0
| Maximum 16K 16K 16K 16K 16K 16K 16K 16K 24K 24K 32K 32K 32K 32K
| devices (UAs)
| per FC channel
| path (Note 10 on
| page 293)
| Channel subsystems (CSSs) and subchannel sets
| Maximum CSS 3 1 3 1 3 1 3 1 3 1 5 2 5 2
| ID (Note 7 on
| page 293)
| Maximum 0 0 1 1 1 1 2 1 2 1 3 2 3 2
| subchannel set ID
| Logical partitions
| Maximum logical 30 30 60 30 60 30 60 30 60 30 85 40 85 40
| partitions (Note 2
| on page 292)
| Maximum I/O 64 64 64 64 64 64 64 64 64 64 128 128 128 128
| Clusters
| Control unit headers (CUHs)
| Maximum CUHs 4096 4096 4096 4096 4096 4096 4096 4096 4096 4096 4096 4096 4096 4096
| per CSS
| Physical control units (PCUs)
| Maximum PCUs 8192 8192 8192 8192 8192 8192 8192 8192 8192 8192 8192 8192 8192 8192
| Channel paths or functions
| Maximum 6FF 2FF 6FF 2FF 6FF 2FF 6FF 2FF 6FF 2FF 37F 2FF 37F 1FF
| PCHID number
| Minimum N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 7C0 7C0 7C0 7C0
| VCHID Number
| Maximum N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 7FF 7FF 7FF 7FF
| VCHID Number
| Minimum AID N/A N/A 0 8 0 0 0 0 0 0 0 0 0 N/A
| number for CIB
| Maximum AID N/A N/A 1F F 1F 5 1F 0B 1F 0B 0F 0F 0F N/A
| number for CIB
| Maximum CIB N/A N/A 8 8 16 6 16 8 16 8 16 8 16 N/A
| AIDs

| Machine Limit 2084,1 2086,1 2094,1 2096,1 2097,1 2098,1 2817,1 2818,1 2827,1 2828,1 2964,1 2965,1 3906,1 3907,1
| Maximum CIB N/A N/A 2 2 2 2 4 4 4 4 4 4 4 N/A
| port value
| Minimum AID N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 10 10 10 10
| number for CS5
| Maximum AID N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 37 37 37 37
| number for CS5
| Maximum CS5 N/A N/A N/A N/A N/A N/A N/A N/A 16 16 20 8 40 8
| AIDs
| Maximum CL5 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 32 16 32 16
| PCHIDs
| Maximum 1024 512 1024 512 1024 512 704 464 528 320 608 448 608 352
| channel paths (Note
| 4 on
| page
| 293)
| Maximum CVC, 1024 420 1024 420 1024 480 240 240 N/A N/A N/A N/A N/A N/A
| CBY, CNC, or (Note
| CTC channel 4 on
| paths page
| 293)
| Maximum FCV, 240 80 336 112 336 128 352 160 368 160 320 160 320 128
| FC, FCP, OSC,
| OSD, OSE, OSM,
| OSN, or OSX
| channel paths
| (Note 3 on page
| 292)
| Maximum FCP, 240 80 256 112 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
| OSC, OSD, OSE,
| or OSN channel
| paths
| Maximum FCP 240 80 256 112 256 128 256 160 256 160 256 160 256 128
| channel paths
| Maximum OSC, 48 40 48 48 48 48 48 48 48 48 48 48 48 48
| OSD, OSE, OSM,
| OSN, or OSX
| channel paths
| Maximum FCV 120 40 168 56 168 64 N/A N/A N/A N/A N/A N/A N/A N/A
| channel paths
| Maximum CFS, 64 64 64 64 64 64 128 128 128 128 256 256 256 224
| CFR, CFP, CBS,
| CBR, CBP, CIB,
| CL5, CS5 or ICP
| channel paths
| Maximum CIB, 64 64 64 64 64 64 128 128 128 128 128 128 128 128
| CL5, CS5, or ICP
| channel paths per
| logical partition

| Machine Limit 2084,1 2086,1 2094,1 2096,1 2097,1 2098,1 2817,1 2818,1 2827,1 2828,1 2964,1 2965,1 3906,1 3907,1
| Maximum CFS, 48 48 48 48 48 48 48 48 48 48 N/A N/A N/A N/A
| CFR, or CFP
| channel paths
| Maximum CFS or 32 32 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
| CFR channel
| paths
| Maximum CBS 8 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
| or CBR channel
| paths
| Maximum CBP 32 16 32 24 16 12 N/A N/A N/A N/A N/A N/A N/A N/A
| channel paths
| Maximum ICP 32 32 32 32 32 32 32 32 32 32 32 32 32 32
| channel paths
| Maximum IQD 16 16 16 16 16 16 32 32 32 32 32 32 32 32
| channel paths
| Maximum N/A N/A N/A N/A N/A N/A N/A N/A 24 16 40 40 56 56
| function adapters
| Maximum N/A N/A N/A N/A N/A N/A N/A N/A FF FF 22FF 22FF 47FF 47FF
| function ID (FID)
| Maximum HYL N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 16 16
| PCHIDs
| Maximum ISM N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 32 32 32 32
| VCHIDs
| Maximum RCE N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 16 16 16 16
| PCHIDs
| Maximum ROCE N/A N/A N/A N/A N/A N/A N/A N/A 16 16 16 16 8 4
| PCHIDs
| Maximum ROC2 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 8 4
| PCHIDs
| Maximum ROCE N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 8 4
| or ROC2 PCHIDs
| Maximum ROCE N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 2 1
| and ROC2
| PCHIDs per
| resource group
| Maximum ZEDC N/A N/A N/A N/A N/A N/A N/A N/A 8 8 8 8 16 8
| PCHIDs
| Maximum VFs N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 63 31
| per ROC2 port
| IOCP storage requirement
| Minimum storage 40 22 74 40 74 40 108 40 108 40 214 86 213 85
| required by IOCP MB* MB* MB* MB* MB* MB* MB* MB* MB* MB* MB* MB* MB* MB*
| *K=1024. MB=1 048 576 bytes.
|

| Table 19. Valid CHPID types
| Machine type (See Note 1 on page 292)
| CHPID
| type 2084,1 2086,1 2094,1 2096,1 2097,1 2098,1 2817,1 2818,1 2827,1 2828,1 2964,1 2965,1 3906,1 3907,1
| CBY Yes Yes Yes Yes Yes Yes Yes Yes No No No No No No
| CVC Yes Yes Yes Yes Yes Yes Yes Yes No No No No No No
| CNC Yes Yes Yes Yes Yes Yes Yes Yes No No No No No No
| CTC Yes Yes Yes Yes Yes Yes Yes Yes No No No No No No
| CFS Yes Yes No No No No No No No No No No No No
| CFR Yes Yes No No No No No No No No No No No No
| CFP Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No No No No
| CBS Yes No No No No No No No No No No No No No
| CBR Yes No No No No No No No No No No No No No
| CBP Yes Yes Yes Yes Yes Yes No No No No No No No No
| (Note 8
| on page
| 293)
| CIB No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No
| (Note 9
| on page
| 293)
| CL5 No No No No No No No No No No Yes Yes Yes Yes
| (Note 12
| on page
| 293)
| CS5 No No No No No No No No No No Yes Yes Yes Yes
| FC Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
| FCV Yes Yes Yes Yes Yes Yes No No No No No No No No
| FCP Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
| (Note 5
| on page
| 293)
| ICP Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
| IQD Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
| OSC Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
| (Note 6
| on page
| 293)
| OSD Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
| OSE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
| OSM No No No No No No Yes Yes Yes Yes Yes Yes Yes Yes
| OSN No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No No
| OSX No No No No No No Yes Yes Yes Yes Yes Yes Yes Yes
|
Machine types not listed do not support function adapters.

| Table 20. Valid FUNCTION types
| Machine type (See Note 1)
| FUNCTION type 2827,1 2828,1 2964,1 2965,1 3906,1 3907,1
| HYL No No No No Yes Yes
| ISM No No Yes Yes Yes Yes
| RCE No No Yes Yes Yes Yes
| ROCE (See Note 11 on page 293) Yes Yes Yes Yes Yes Yes
| ROC2 No No No No Yes Yes
| ZEDC (See Note 11 on page 293) Yes Yes Yes Yes Yes Yes
|
Notes:
1. Machine type refers to the value specified in the SYSTEM keyword of the ID statement.
v SYSTEM=(2084,1) specifies a 2084 zSeries 990 Model A08, B16, C24, or D32.
v SYSTEM=(2086,1) specifies a 2086 zSeries 890 Model A04.
The smallest sub-uniprocessor equipped capacity setting has additional restrictions IOCP does not
enforce (for example, a maximum of 240 ESCON channels, 24 OSA channels, and 15 logical
partitions).
v SYSTEM=(2094,1) specifies a 2094 Model S08, S18, S28, S38, or S54.
v SYSTEM=(2096,1) specifies a 2096 Model R07 or S07.
v SYSTEM=(2097,1) specifies a 2097 Model E12, E20, E40, E56, or E64.
v SYSTEM=(2098,1) specifies a 2098 Model E10.
v SYSTEM=(2817,1) specifies a 2817 Model M15, M32, M49, M66, or M80.
v SYSTEM=(2818,1) specifies a 2818 zEnterprise 114 Model M05 or M10.
v SYSTEM=(2827,1) specifies a 2827 zEnterprise EC12 Models H20, H43, H66, H89, or HA1.
v SYSTEM=(2828,1) specifies a 2828 zEnterprise BC12 Models H06 or H13.
v SYSTEM=(2964,1) specifies a 2964 Model N30, N63, N96, NC9, NE1, L30, L63, L96, LC9, or LE1.
v SYSTEM=(2965,1) specifies a 2965 Model N10, N20, L10, or L20.
v SYSTEM=(3906,1) specifies a 3906 Model M01, M02, M03, M04, M05, LM1, LM2, LM3, LM4, or
LM5.
| v SYSTEM=(3907,1) specifies a 3907 Model ZR1 or LR1.
2. The 2084 models support 15 logical partitions at EC J12560 and 30 logical partitions at EC J12560
(with an LMC patch installed) or later.
The 2096 Model R07 supports the activation of only 15 logical partitions (LPs) on the system. IOCP
issues warning message ICP176I if an IOCDS containing more than 15 LPs is written to a Model R07.
You can power-on reset with an IOCDS containing more than 15 LPs but only 15 of them can be
activated.
3. You are informed if you define more than the maximum number of channel paths. This maximum
number of channel paths is determined by the maximum number of cards that can be installed in
the processor and can be reached only if every card is a FICON card. Also, this maximum includes
PCI Cryptographic and Flash Express cards installed on your CPC. However, PCI Cryptographic and
Flash Express cards are not defined to IOCP, and IOCP cannot include them in its checking of this
maximum. Also, function adapter cards (for example, zEDC Express) use FUNCTION statements for
their definition instead of CHPID statements and are excluded from the maximum channel path
check. The following table summarizes the maximum for each type of CPC:

Table 21. Maximums by CPC
CPC Maximum number of cards
| 3907 64
3906 160
2965 80
2964 160
2828 80
2827 168
2818 72
2817 168
2098 32
2097 84
2096 28
2094 84
2086 20
2084 60
4. RPQs might be available to increase these values.

5. The 2084 models support FCP channel paths at EC J12560 (with an LMC patch installed) or later.
6. The 2084 models support OSC channel paths at EC J13486.
7. The 2084 models support a maximum CSS ID of 1 at EC J12560 and a maximum CSS ID of 3 at EC
J13486.
8. CBP channel paths are supported by all 2097 models except E64.
9. CIB channel paths are supported as follows:
v On 2094 models at EC G40965 (with an LMC patch installed) or later
v On 2096 models at EC G40965 (with an LMC patch installed) or later
v On 2097 models at EC F85906 (with an LMC patch installed) or later
v On all ECs of any other CPC that supports the CIB channel path type.
10. IOCP might generate more unit address ranges and unit addresses for a spanned FC channel path
than for an unspanned channel path. See “FC control unit definition considerations:” on page 86.
The 2827 models support 16K devices (UAs) at EC H09173 and 24K devices (UAs) at EC H49564.
11. The 2827 and 2828 models support RoCE and zEDC function adapters but ignore the FUNCTION
TYPE keyword.
12. CL5 channel paths are supported as follows:
v On 2965 and 2964 models at EC P00339 (with an LMC patch installed)
v On all ECs of any other CPC that supports the CL5 channel path type.
Table 22. FCP Channel Path Limits
Limit or Recommendation FICON Express 8/8S FICON Express 16S/16S+
Maximum configurable devices or 480 480
subchannels per FCP channel path
Recommended maximum number of 32 64
active NPIV-enabled subchannels per
FCP channel path
Maximum number of remote N_Ports 512 1024
per FCP channel path

Table 22. FCP Channel Path Limits (continued)
Limit or Recommendation FICON Express 8/8S FICON Express 16S/16S+
Maximum concurrent I/O operations 960 1528
per FCP channel path
Maximum number of LUNs that can 4096 8192
be open at the same time
Keyword applicability for channel paths and functions

Keyword applicability lists the keyword applicability for the channel path types and function types that
IOCP supports.
Table 23. Keyword applicability for channel paths
Channel path type
OSC
OSD
OSE
CNC CFP OSM
CTC CBP CFS CIB CFR OSN
Keyword CVC CBY FCV ICP CBS CS5 CL5 CBR OSX FC IQD FCP
CHPID
PCHID Yes Yes Yes Yes (1) Yes No Yes Yes Yes Yes No Yes
VCHID No No No No No No No No No No Yes(8) No
PATH to span chpid No No No Yes (5) Yes (5) Yes Yes No Yes (5) Yes (5) Yes (5) Yes (5)
SWITCH Yes Yes Yes No No No No No No Yes No No
NOTPART No No Yes Yes Yes Yes Yes No Yes Yes Yes Yes
SHARED No No Yes Yes Yes Yes Yes No Yes Yes Yes Yes
CPATH No No No Yes (3) No Yes Yes No No No No No
CHPARM No No Yes (4) No No No No No Yes (4) Yes Yes No
AID No No No No No Yes No No No No No No
CSYSTEM No No No No No Yes Yes No No No No No
PORT No No No Yes No Yes Yes No No No No No
PNETID No No No No No No No No Yes (7) No Yes No
CNTLUNIT
LINK Yes Yes Yes No No No No N/A No Yes No No
CUADD No No Yes No No No No N/A Yes (6) Yes Yes No
PROTOCL Yes No No No No No No N/A No No No No
SHARED Yes No No No No No No N/A No No No No
UNITADD Yes Yes Yes No No No No N/A No Yes No No
IODEVICE
PART, PARTITION, No No Yes No No No No N/A Yes Yes Yes Yes
or NOTPART
PATH Yes Yes Yes (2) No No No No N/A No Yes No No
SCHSET No No Yes No No No No N/A Yes Yes Yes Yes
STADET Yes Yes Yes No No No No N/A No Yes No No
TIMEOUT Yes No No No No No No N/A No No No No

Table 23. Keyword applicability for channel paths (continued)
Channel path type
OSC
OSD
OSE
CNC CFP OSM
CTC CBP CFS CIB CFR OSN
Keyword CVC CBY FCV ICP CBS CS5 CL5 CBR OSX FC IQD FCP
UNITADD Yes Yes Yes No No No No N/A Yes Yes Yes Yes
1. The PCHID keyword is not allowed for ICP channel paths.

2. The PATH keyword on the IODEVICE statement is ignored for CTC channel paths.
3. The CPATH keyword is not allowed for CFP and CBP channel paths.
4. The CHPARM keyword is allowed for CNC, FCV, OSC, OSD, OSM, and OSX channel paths.
5. ICP and IQD channel paths can be spanned on 2084 models at EC J12560 (with an LMC patch installed) or later
and all other models. FC, FCP, CFS, CBS, CFP, CBP, OSC, OSD, and OSE channel paths can be spanned on 2084
models at EC J13486 and all other models. OSM, OSN, and OSX channel paths can be spanned on all models
that support the channel path types.
6. The CUADD keyword is allowed for OSD, OSM, and OSX channel paths.
7. The PNETID keyword is allowed for OSD channel paths on 2827, 2828, and later models.
8. The VCHID keyword is allowed for IQD channel paths on the 2964, 2965, and later models.
Keyword applicability lists the keyword applicability for the function types that IOCP supports.
Table 24. Keyword applicability for functions
Function type
Keyword HYL ISM RCE ROCE ROC2 ZEDC
FUNCTION
PCHID Yes No Yes Yes Yes Yes
VCHID No Yes No No No No
PNETID No Yes No Yes Yes No
PORT Yes No No No Yes No
VF Yes Yes No Yes Yes Yes
Machine types not listed do not support the keyword or statement.

| Table 25. Statement and keyword applicability for machine types
| Statement or Keyword 2827,1 2828,1 2964,1 2965,1 3906,1 3907,1
| UUID statement No No Yes (1) Yes (1) Yes Yes
| FUNCTION statement Yes Yes Yes Yes Yes Yes
| UID keyword on FUNCTION No No Yes (1) Yes (1) Yes Yes
| statement
| VCHID on CHPID statement No No Yes Yes Yes Yes
| TYPE keyword on FUNCTION No No Yes Yes Yes Yes
| statement
| 1. The UUID statement and UID keyword on FUNCTION statement are supported on 2965 and 2964 models at
| P00339 (with an LMC patch installed).
|

CPC activation and power-on reset error
CPCs can fail activation or power-on reset (POR) if the selected IOCDS contains an unsupported function
or channel path type for the CPC. This situation can occur if your CPC does not yet support a function
IOCP permits. The activation or power-on reset fails with error message (ACTZ16F) as follows:
The requested operation cannot be performed because some element of the
configuration defined in the input/output configuration data set (IOCDS)
is not supported. xxxxxxxx xxxxxxxx indicates the unsupported function or
channel type.
See section ’CPC Activation and power-on reset Error’ in the ’Machine
Limits and Rules’ appendix in the IOCP User’s Guide for the supported
functions and channel types. Verify that the IOCDS definition is correct for
this machine EC level.
Contact IBM support for further help.
This CPC error message contains a hexadecimal field that identifies the unsupported function or channel
path type. Use the following table to determine which function or channel path type is incorrect for your
CPC and write a new IOCDS that does not contain it. Also, the following hexadecimal values describe
what the displays when you click Supported I/O Mask under View on the . If there is a mismatch
between the Supported I/O mask and the IOCDS mask that the SE displays, an Unsupported mask is
displayed that contains only the unsupported functions or channel path types. Also, power-on reset fails
with the IOCDS at the current EC.
IOCP message ICP065I also contains a hexadecimal field which identifies unsupported functions or
channel path types. The following table lists hexadecimal values and their meanings:
Table 26. Activation and POR error message hexadecimal values
Hexadecimal value Channel path type, function type, or IOCP function
00000000 00000001 OSM
00000000 00000002 OSX
00000000 00000004 CNC
00000000 00000008 CTC
00000000 00000010 CVC
00000000 00000020 CBY
00000000 00000080 CFS
00000000 00000100 CFR
00000000 00000200 CIB
00000000 00001000 CBS
00000000 00002000 CBR
00000000 00004000 FCV
00000000 00008000 FC
00000000 00020000 OSN
00000000 00080000 OSD
00000000 00100000 OSE
00000000 00200000 OSC
00000000 00400000 CFP
00000000 00800000 CBP
00000000 01000000 ICP
00000000 02000000 FCP

Table 26. Activation and POR error message hexadecimal values (continued)
Hexadecimal value Channel path type, function type, or IOCP function
00000000 04000000 IQD
00000000 10000000 CS5
00000000 20000000 CL5
00000000 80000000 FUNCTION FID defined (Note 5)
00000001 00000000 ROCE FID
00000002 00000000 ZEDC FID
00000008 00000000 ISM FID
00000010 00000000 RCE FID
00000020 00000000 HYL FID
00000080 00000000 ROC2 FID
01000000 00000000 CHECKCPC=NO or NOCHKCPC
04000000 00000000 UUID statement specified (Note 7)
08000000 00000000 FC channel path defined with greater than 16384 unit addresses on CNTLUNIT
statements (Note 6)
10000000 00000000 PNETID keyword specified (Note 6)
20000000 00000000 OSD channel path defined with nonzero CHPARM or OSD control unit defined with
nonzero CUADD (Notes 1, 4)
40000000 00000000 CSS ID > 1 or spanned FC, FCP, CFS, CBS, CFP, CBP, OSC, OSD, or OSE channel path
defined (Notes 1, 3)
80000000 00000000 Spanned ICP or IQD channel path (Notes 1, 2)
Notes:
1. This value is set only for 2084 models.
2. This IOCP function is supported on 2084 models at EC J12560 (with an LMC patch installed) or later.
3. This IOCP function is supported on 2084 models at EC J13486.
4. This IOCP function is supported on 2084 models at EC J13486 (with an LMC patch installed)
5. This value is set only for 2828 and 2827 models.
6. This value is set only for 2877 models.
7. This value is set only for 2965 and 2964 models.

Appendix G. Notices
This information was developed for products and services offered in the USA.
IBM may not offer the products, services, or features discussed in this document in other countries.
Consult your local IBM representative for information on the products and services currently available in
your area. Any reference to an IBM product, program, or service is not intended to state or imply that
only that IBM product, program, or service may be used. Any functionally equivalent product, program,
or service that does not infringe any IBM intellectual property right may be used instead. However, it is
the user's responsibility to evaluate and verify the operation of any non-IBM product, program, or
service.
IBM may have patents or pending patent applications covering subject matter described in this
document. The furnishing of this document does not grant you any license to these patents. You can send
license inquiries, in writing, to:
IBM Director of Licensing

IBM Corporation
North Castle Drive
Armonk, NY 10504-1785 USA
INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS PUBLICATION “ AS IS”

WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR
FITNESS FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer of express or implied
warranties in certain transactions, therefore, this statement may not apply to you.
This information could include technical inaccuracies or typographical errors. Changes are periodically
made to the information herein; these changes will be incorporated in new editions of the publication.
IBM may make improvements and/or changes in the product(s) and/or the program(s) described in this
publication at any time without notice.
Any references in this information to non-IBM websites are provided for convenience only and do not in
any manner serve as an endorsement of those websites. The materials at those websites are not part of
the materials for this IBM product and use of those websites is at your own risk.
IBM may use or distribute any of the information you supply in any way it believes appropriate without
incurring any obligation to you.
Any performance data contained herein was determined in a controlled environment. Therefore, the
results obtained in other operating environments may vary significantly. Some measurements may have
been made on development-level systems and there is no guarantee that these measurements will be the
same on generally available systems. Furthermore, some measurements may have been estimated through
extrapolation. Actual results may vary. Users of this document should verify the applicable data for their
specific environment.
Information concerning non-IBM products was obtained from the suppliers of those products, their
published announcements or other publicly available sources. IBM has not tested those products and
cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM
products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of
those products.

All statements regarding IBM's future direction or intent are subject to change or withdrawal without
notice, and represent goals and objectives only.
All IBM prices shown are IBM's suggested retail prices, are current and are subject to change without
notice. Dealer prices may vary.
This information is for planning purposes only. The information herein is subject to change before the
products described become available.
This information contains examples of data and reports used in daily business operations. To illustrate
them as completely as possible, the examples include the names of individuals, companies, brands, and
products. All of these names are fictitious and any similarity to the names and addresses used by an
actual business enterprise is entirely coincidental.
If you are viewing this information softcopy, the photographs and color illustrations may not appear.
Trademarks
IBM, the IBM logo, and ibm.com® are trademarks of International Business Machines Corp., registered in
many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other
companies. A current list of IBM trademarks is available on the web at “Copyright and trademark
information” at www.ibm.com/legal/copytrade.shtml.
Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both.
Microsoft, Windows, and Windows logo the are trademarks of Microsoft Corporation in the United
States, other countries, or both.
Class A Notices
The following Class A statements apply to this IBM product. The statement for other IBM products
intended for use with this product will appear in their accompanying manuals.
Federal Communications Commission (FCC) Statement
Note: This equipment has been tested and found to comply with the limits for a Class A digital device,
pursuant to Part 15 of the FCC Rules. These limits are designed to provide reasonable protection against
harmful interference when the equipment is operated in a commercial environment. This equipment
generates, uses, and can radiate radio frequency energy and, if not installed and used in accordance with
the instruction manual, may cause harmful interference to radio communications. Operation of this
equipment in a residential area is likely to cause harmful interference, in which case the user will be
required to correct the interference at his own expense.
Properly shielded and grounded cables and connectors must be used in order to meet FCC emission
limits. IBM is not responsible for any radio or television interference caused by using other than
recommended cables and connectors or by unauthorized changes or modifications to this equipment.
Unauthorized changes or modifications could void the user's authority to operate the equipment.
This device complies with Part 15 of the FCC rules. Operation is subject to the following two conditions:
(1) this device may not cause harmful interference, and (2) this device must accept any interference
received, including interference that may cause undesired operation.
Industry Canada Compliance Statement
This Class A digital apparatus complies with Canadian ICES-003.

Avis de conformité à la réglementation d'Industrie Canada
Cet appareil numérique de la classe A est conforme à la norme NMB-003 du Canada.
European Community Compliance Statement
This product is in conformity with the protection requirements of EU Council Directive 2014/30/EU on
the approximation of the laws of the Member States relating to electromagnetic compatibility. IBM cannot
accept responsibility for any failure to satisfy the protection requirements resulting from a
non-recommended modification of the product, including the fitting of non-IBM option cards.
This product has been tested and found to comply with the limits for Class A Information Technology
Equipment according to European Standard EN 55032. The limits for Class A equipment were derived for
commercial and industrial environments to provide reasonable protection against interference with
licensed communication equipment.
European Community contact:

IBM Deutschland GmbH
Technical Regulations, Department M372
IBM-Allee 1, 71139 Ehningen, Germany
Tele: +49 (0) 800 225 5423 or +49 (0) 180 331 3233
email: [email protected]
Warning: This is a Class A product. In a domestic environment, this product may cause radio
interference, in which case the user may be required to take adequate measures.
VCCI Statement - Japan
The following is a summary of the Japanese VCCI statement above:
This is a Class A product based on the standard of the VCCI Council. If this equipment is used in a
domestic environment, radio interference may occur, in which case the user may be required to take
corrective actions.
Japan JIS C 61000-3-2 Compliance
（"#）%&'()*+,-./ 012%3456789
/ :;に=づく@ABC%CD： Knowledge CenterのGHIの/
/ / / / / / / / / / / / / JKページOP
For products less than or equal to 20 A per phase, the following statement applies:
!"#$%&' JIS C 61000-3-2 ()*
Appendix G. Notices 301

For products greater than 20 A, single-phase, the following statements apply:
,:;3TUV JIS C 61000-3-2 WXY
&'(は、「,-.は/0,-で23する678の,:;<=>
?ガイドライン」>FGH（,:;JKGH）です。
#$NO：６（!"、ＰＦＣ#$%）
PQRS：０
For products greater than 20 A per phase, three-phase, the following statements apply:
%&'()*+ JIS C 61000-3-2 ,-.
/01は、「%56は78%5で:(する=>?の%&'ABC
Dガイドライン」CKLM（%&'OPLM）です。
"#ST ：５ (３!、ＰＦＣ"#$）
UVWX ：０
Electromagnetic Interference (EMI) Statement - People's Republic of China
Declaration: This is a Class A product. In a domestic environment, this product may cause radio
interference, in which case the user may need to perform practical action.
Electromagnetic Interference (EMI) Statement - Taiwan
The following is a summary of the Taiwan EMI statement above:

Warning: This is a Class A product. In a domestic environment, this product may cause radio
interference, in which case the user will be required to take adequate measures.
IBM Taiwan Contact Information:
Electromagnetic Interference (EMI) Statement - Korea
! ""# $%&( A')() *+,-./01 2 ""!34

56+ 7# 8&+# ! 91 :;<=" >?@, ABC;
DEFG 8&<# H1 I-() .4J.
Germany Compliance Statement

Deutschsprachiger EU Hinweis: Hinweis für Geräte der Klasse A EU-Richtlinie zur
Elektromagnetischen Verträglichkeit
Dieses Produkt entspricht den Schutzanforderungen der EU-Richtlinie 2014/30/EU zur Angleichung der
Rechtsvorschriften über die elektromagnetische Verträglichkeit in den EU-Mitgliedsstaaten und hält die
Grenzwerte der EN 55032 Klasse A ein.
Um dieses sicherzustellen, sind die Geräte wie in den Handbüchern beschrieben zu installieren und zu
betreiben. Des Weiteren dürfen auch nur von der IBM empfohlene Kabel angeschlossen werden. IBM
übernimmt keine Verantwortung für die Einhaltung der Schutzanforderungen, wenn das Produkt ohne
Zustimmung von IBM verändert bzw. wenn Erweiterungskomponenten von Fremdherstellern ohne
Empfehlung von IBM gesteckt/eingebaut werden.
EN 55032 Klasse A Geräte müssen mit folgendem Warnhinweis versehen werden:
"Warnung: Dieses ist eine Einrichtung der Klasse A. Diese Einrichtung kann im Wohnbereich
Funk-Störungen verursachen; in diesem Fall kann vom Betreiber verlangt werden, angemessene
Maßnahmen zu ergreifen und dafür aufzukommen."
Deutschland: Einhaltung des Gesetzes über die elektromagnetische Verträglichkeit von Geräten
Dieses Produkt entspricht dem “Gesetz über die elektromagnetische Verträglichkeit von Geräten
(EMVG)“. Dies ist die Umsetzung der EU-Richtlinie 2014/30/EU in der Bundesrepublik Deutschland.
Zulassungsbescheinigung laut dem Deutschen Gesetz über die elektromagnetische Verträglichkeit von
Geräten (EMVG) (bzw. der EMC EG Richtlinie 2014/30/EU) für Geräte der Klasse A
Dieses Gerät ist berechtigt, in Übereinstimmung mit dem Deutschen EMVG das EG-Konformitätszeichen
- CE - zu führen.
Verantwortlich für die Einhaltung der EMV Vorschriften ist der Hersteller:
International Business Machines Corp.
New Orchard Road
Armonk, New York 10504
Tel: 914-499-1900
Appendix G. Notices 303

Der verantwortliche Ansprechpartner des Herstellers in der EU ist:
IBM Deutschland GmbH
Technical Regulations, Abteilung M372
IBM-Allee 1, 71139 Ehningen, Germany
Tel: +49 (0) 800 225 5423 or +49 (0) 180 331 3233
email: [email protected]
Generelle Informationen:
Das Gerät erfüllt die Schutzanforderungen nach EN 55024 und EN 55032 Klasse A.
Electromagnetic Interference (EMI) Statement - Russia

Index
Special characters CHPID statement (continued)
introduction 20, 22
CSS keyword parameter (continued)
in PART or PARTITION keyword
* IOCP comment card 237 CIB keyword value 67 (continued)
*$HCD$ 111, 119 CL5 keyword value 67 in RESOURCE statement 60
*$HCDC$ 111, 119 CMS/DOS 124 in PATH keyword
** and **** for link addresses 84 CNC keyword value 66 in CHPID statement 63
*IOCP comment card 53 CNTLUNIT statement in CNTLUNIT statement 82
correspondence with HCPRIO file 55 in IODEVICE statement 98
description 80 CSYSTEM keyword 78
Numerics format 82 CTC keyword value 66
9035 ESCON converter introduction 20, 22 CUADD keyword 86
considerations 43 keyword values 245 CUNUMBR keyword
codes in CNTLUNIT statement 82
return 116, 127, 136 in IODEVICE statement 94
A wait state 148
comment cards, IOCP 53, 237
customer name for IOCDS 58
CVC keyword value 66
A keyword value CONFIG frame 21
in REPORT option 115, 135 configuration considerations 21
access list 71, 73, 105
accessibility xix
configuration data
initially defining 117, 128
D
contact IBM xix data streaming protocol
introduction 1
features xx 3 MB/sec 88
redefining 118, 129
ADDRESS keyword 92 4.5 MB/sec 88
specifying 19
AID keyword 78 DC interlock protocol
updating 118, 129
assistive technologies xix in PROTOCL keyword 88
configuration reports
asterisks for link addresses 84 defining
conditions for producing 170
Ax keyword value IQD channels 37
description 151
in REPORT option 115, 135 device candidate list 96
heading, ID statement 58
in WRTCDS option 113, 132 DISK option in IOCP command 123
JCL example 115, 135
DMSICP messages 233
producing 110, 123
DYN option in IOCP command 124
console 20
C continuation 238
candidate list 72, 74 control unit headers 22
candidate list names 105 control unit types 45 E
card-image input control units ESA/390 mode
introduction 20 examples 24 channel path selection
IOCP statements 53 introduction 1 preferred path 44
CBP keyword value 67 restrictions 23 control unit headers 22
CBR keyword value 67 specifying in CNTLUNIT control unit types 45
CBS keyword value 67 statement 80, 81, 245 ESCON channel path 66
CBY keyword value 65 types 45 examples of statement lists 241
CFP keyword value 66 coupling facility 27
CFR keyword value 66 coupling facility channels 27
CFS keyword value 67 CPATH keyword 75
CPC modes and the IOCDS 21
F
channel path selection FC keyword value 69
for ESA/390 mode 44 create IOCDS for channel subsystem 21
FCP channels 37
preferred path 44 CS5 keyword value 68
FCP keyword value 70
channel paths CSS ID 60, 62
FCV keyword value 69
introduction 1 CSS keyword parameter
Fibre Channel Protocol (FCP)
specifying in CHPID statement 64 in CPATH keyword
support 37
channel subsystem 21 in CHPID statement 75
FICON channels 69
CHECKCPC option in LINK keyword
FID keyword 104
in IOCP command 122 in CNTLUNIT statement 84
file ID 120
in PARM parameter of EXEC in MAXDEV keyword
FILEDEF command 120, 125
statement 113, 132 in RESOURCE statement 61
format of statements
CHPARM keyword 76 in NOTPART keyword
CHPID 64
chpid number 64 in CHPID statement 73
CNTLUNIT statement 82
CHPID statement in IODEVICE statement 96
FUNCTION 104
correspondence with HCPRIO file 55 in PART or PARTITION keyword
ID 58
description 63 in CHPID statement 71
IODEVICE 92
format 64 in IODEVICE statement 96
RESOURCE 60

format of statements (continued) in FUNCTION statement (continued) IOCP configuration report 151
rules for coding 237 TYPE keyword 106 IOCP messages 171
UUID 102 input to IOCP IOCP statements and functions 53
FUNCTION statement introduction 20 IODEVICE statement
examples 108 IOCP statements 53 correspondence with HCPRIO file 55
format 104 input/output configuration data set 20 description 92
introduction 20 input/output configuration program. 1 examples 101
Input/Output Configuration window of format 92
the Support Element 21 introduction 20
G integrated cluster bus channels 27, 67
internal coupling channels 27
keyword values 245
IQD channels, defining 37
generating
introduction to IOCP 1 IQD keyword value 70
VSE version 49
IOCDS ISM keyword value 106
generating an IOCDS
configuration considerations 21
overview 20
configuration reports 109, 151
stand-alone version 50, 144
VM version 47
CPC modes 21
customer name 58
J
examples 126 JCL statements
generating
VSE version 49 z/OS version
stand-alone version 144
z/OS version 46 generating an IOCDS 112
VM version 47
JCL examples 114 introduction 46, 111
z/OS version 114
z/VSE version producing configuration
initially defining 117, 128, 137, 139,
JCL examples 133 reports 115
144
z/VSE version
overview 20
generating an IOCDS 131
redefining 118, 129, 137, 146
H write-protection 21, 143
introduction 131
producing configuration
hardware configuration definition 46 IOCLUSTER keyword 78
reports 135
HCD IOCP
VM version 48, 119 codes 116, 127, 136, 148
z/OS version 46, 111 comment card 53
HCPRIO file 55 configuration considerations 21 K
HYL keyword value 106 configuration reports 151 keyboard
CPC modes 21 navigation xix
introduction 1 keyword values
I keyword values 245
messages
in IODEVICE and CNTLUNIT
keywords
245
I/O configuration data
about VM only 233 in CHPID statement
specifying 19, 53
about z/OS, VM, and VSE 171 CHPARM 76
I/O configuration data set 1
return codes 116, 127, 136 CPATH 75
I/O devices
running VM version 119 CSYSTEM 78
introduction 1
running z/OS version 111 IOCLUSTER 78
specifying in IODEVICE
specifying statements 53 MIXTYPE 70
statement 92, 245
statements 53 NOTPART 73
I/O interface timeout function 243
version and release number on the PART 71
I/O interface timeout function. 243
reports 151 PARTITION 71
I/O subsystem 1
versions 45 PATH 64
ICP
wait state codes 148 PCHID 65
generating an IOCDS
IOCP command PORT 78
CHECKCPC option 122 SHARED 75
VM version 126
DISK option 123 SWITCH 70
z/OS version 46
DYN option 124 TYPE 65
in CPATH keyword
general usage notes 125 VCHID 65
in CHPID statement 75
ICP option 121 in CNTLUNIT statement
in PATH keyword
LINECOUN option 123 CUADD 86
in IODEVICE statement 98
NOCHKCPC option 123 CUNUMBR 82
keyword value 67
NODYN option 124 LINK 84
option in IOCP command 122
NOPRINT option 124 PATH 82
ID statement
NOTERM option 124 PROTOCL 88
description 58
NOWARN option 124 SHARED 88
example 59
NOWRTCDS option 122 UNIT 89
format 58
PRINT option 124 UNITADD 89
in configuration reports 151
RPTA option 123 in FUNCTION statement
introduction 20
RPTAx option 123 FID 104
ID1=, ID2= 58
TERMINAL option 124 PART 105
in IOCP reports 151
WARNING option 124 PCHID 104
identification information 58
WRTAx option 122 PNETID 107
in FUNCTION statement
IOCP command general usage notes 125 PORT 105
PORT keyword 105

keywords (continued) MSG1 keyword PART keyword (continued)
in FUNCTION statement (continued) in ID statement 58 in FUNCTION statement 105
TYPE 106 MSG2 keyword in IODEVICE statement 96
UID 107 in ID statement 58 in RESOURCE statement 60
VCHID 104 in UUID statement 103
VF 106 PARTITION keyword
in ID statement
MSG1 58
N in CHPID statement 71
N keyword value
MSG2 58 in RESOURCE statement 60
in SHARED keyword 88
SYSTEM 59 in UUID statement 103
in STADET 99
TOK 59 PATH keyword
in TIMEOUT keyword 99
in IODEVICE statement in CHPID statement 64
name 60, 71, 74, 96
ADDRESS 92 in CNTLUNIT statement 82
navigation
CUNUMBR 94 in IODEVICE statement 98
keyboard xix
MODEL 96 PCHID keyword
NO keyword value
NOTPART 96 in CHPID statement 65
in CHECKCPC option 114, 132
PART 96 in FUNCTION statement 104
in WRTCDS option 113, 132
PARTITION 96 PCHID report 17
NOCHKCPC option in IOCP
PATH 98 from the Order Process
command 123
STADET 99 Configurator 17
NODYN option in IOCP command 124
TIMEOUT 99 PCIe adapter
NOPRINT option in IOCP
UNIT 100 introduction 1
command 124
UNITADD 100 PNETID keyword
NOTERM option in IOCP command 124
in RESOURCE statement in CHPID statement 75
NOTPART keyword
MAXDEV 61 in FUNCTION statement 107
PART 60 PORT keyword
PARTITION 60 in CHPID statement 78
NOWARN option in IOCP
in UUID statement in FUNCTION statement 105
command 124
PART 103 power-on reset 46, 48, 51
NOWRTCDS option in IOCP
PARTITION 103 preferred path
command 122
introduction 44
L O
PRINT option in IOCP command 124
privilege class 121
LINECOUN option in IOCP
open systems adapter 31 program options
command 123
open systems adapter Support purpose 121
LINECOUNT option
Facility 32 PROTOCL keyword
in PARM parameter of EXEC
options, program 121 in CNTLUNIT statement 88
statement 113, 115, 132, 135
Order Process Configurator 17
LINK keyword 84
OSA 31
in CNTLUNIT statement 84
lists of statements 241
OSA channels 32
OSA Support Facility
R
logical channel subsystem (CSS) 21 RCE keyword value 106
for VSE/ESA 32
logical path management 41 REC keyword value
for z/OS 32
LPAR in PARTITION keyword 73
OSA-ICC 32
mode 21 REPORT option
OSC channels 32, 33, 34
LSYSTEM keyword 59 condition for coding 116, 135
OSC keyword value 68
OSD channels 32, 33
statement 115, 135
OSD keyword value 68
M OSE channels 33
reports. 110
reserved logical partition 61
machine limits and rules 285 OSE keyword value 68
RESOURCE statement
managed channel 76 OSM channels 33
description 60
MAXDEV keyword 61 OSM keyword value 68
format 60
messages, IOCP OSN channels 33
introduction 21
about VM only 233 OSN keyword value 69
return codes 116, 127, 136
z/OS, z/VM, and z/VSE 171 OSX channels 33
ROC2 keyword value 107
methods of running IOCP OSX keyword value 69
ROCE|ROC keyword value 107
RPTA option in IOCP command 123
VM version 128
RPTAx option in IOCP command 123
z/OS version 117
z/VSE version 136
P rules for coding statements 237
PARM options running IOCP
MIN keyword value 62
generating an IOCDS 111, 131 stand-alone version 50, 139
MIXTYPE keyword 70
in EXEC statement 46, 111, 131 VM version 47, 119
MODEL keyword
producing configuration reports 115, z/OS version 46, 111
135 z/VSE version 48
modes and the IOCDS 21
PART keyword
Index 307
S syntax of statements (continued)
RESOURCE 60
WRTCDS option
sample lists of statements 241 rules for coding 237 statement
SCHSET keyword SYSIN FILEDEF 120 for z/OS 112
in IODEVICE statement 98 SYSPRINT FILEDEF 120, 125 for z/VSE 131
SE system console
with stand-alone version of IOCP input to IOCP 20
overview 50, 139
running IOCP 141, 142
running IOCP 19
starting IOCP 50
Y
with VM version of IOCP Y keyword value
system generation, VM
generating an IOCDS 126 in SHARED keyword 88
relationship to IOCP 47
overview 47 in STADET keyword 99
statements
with z/OS version of IOCP in TIMEOUT keyword 99
RDEVICE 55
generating an IOCDS 114 YES keyword value
SYSTEM keyword 59
overview 46 in CHECKCPC option 113, 132
with z/VSE version of IOCP
generating an IOCDS 133
SE (Support Element) T Z
with stand-alone version of IOCP TERMINAL option in IOCP
z/OS version of IOCP
overview 139 command 124
JCL statements 112, 115
service representative 128, 137, 140, 145 timeout function 243
list examples 241
shared channel 75 TIMEOUT keyword
messages 171
SHARED keyword in IODEVICE statement 99, 243
methods of running 117
in CHPID statement 75 TOK keyword 59
overview 46, 54
in CNTLUNIT statement 88 TYPE keyword
using 111
shortcut keys xix in CHPID statement 65
z/VM version of IOCP
spanned channel 64 in FUNCTION statement 106
messages 171, 233
specifying I/O configuration data 19 overview 54
specifying IOCP statements 53 z/VSE version of IOCP
STADET keyword 99 U JCL statements 135
stand-alone version of IOCP UID keyword 107 messages 171
overview 50 UNIT keyword methods of running 136
statement in CNTLUNIT statement 89 overview 48, 55
IOCP in IODEVICE statement 100 ZEDC|EDC keyword value 107
CHPID 63 UNITADD keyword
CNTLUNIT 80 in CNTLUNIT statement 89
FUNCTION 103 in IODEVICE statement 100
ID 58 UUID statement
IODEVICE 92 examples 103
RESOURCE 60 format 102
UUID 102 introduction 20
list of input 241
rules for coding 237
status verification facility 99
storage requirements for IOCP 112, 119 V
subchannel set 21, 98 VCHID keyword 65
Support Element in CHPID statement 65
with stand-alone version of IOCP in FUNCTION statement 104
running IOCP 142 versions of IOCP 45
with VM version of IOCP VF keyword 106
overview 47 VM version of IOCP
Support Element (SE) 48 methods of running 128
with stand-alone version of IOCP overview 47
overview 50 using 119
running IOCP 141
with VM version of IOCP
generating an IOCDS 126 W
with z/OS version of IOCP wait state codes 148
generating an IOCDS 114 WARNING option in IOCP
overview 46 command 124
with z/VSE version of IOCP with z/VSE version of IOCP
generating an IOCDS 133 overview 48
SWITCH keyword 70 write-protection 21, 143
syntax of statements WRTAx option in IOCP command 122
CHPID 63
CNTLUNIT 80
ID 58
IODEVICE 92

IBM®
Printed in USA
SB10-7172-01

Input/Output Configuration Program User's Guide For ICP Iocp

Uploaded by

Copyright:

Available Formats

Input/Output Configuration Program User's Guide For ICP Iocp

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Input/Output Configuration Program User's Guide For ICP Iocp

Uploaded by

Copyright:

Available Formats

IBM

© Copyright IBM Corp. 2017, 2018 iii

Chapter 6. Running the stand-alone

iv IOCP User's Guide

© Copyright IBM Corp. 2017, 2018 vii

© Copyright IBM Corp. 2017, 2018 ix

World trade safety information

Laser safety information

© Copyright IBM Corp. 2017, 2018 xi

© Copyright IBM Corp. 2017, 2018 xiii

| z14 / Rockhopper II: Models ZR1 and LR1

z13s / Rockhopper: Models N10, N20, L10, and L20

zEnterprise BC12 - 2828: Models H06 and H13

zEnterprise 114 - 2818: Models M05 and M10

z10 BC - 2098: Model E10

z10 EC - 2097: Models E12, E26, E40, E56, and E64

z9 EC - 2094: Models S08, S18, S28, S38, and S54

z9 BC - 2096: Models R07 and S07

z990 - 2084: Models A08, B16, C24, and D32

z890 - 2086: Model A04

What is included in this user's guide

xiv IOCP User's Guide

The appendixes in this user's guide include:

z/OS and z/OS.e

About this publication xv

xvi IOCP User's Guide

About this publication xvii

xviii IOCP User's Guide

Enterprise systems connection (ESCON)

Fibre connection (FICON)

Open Systems Adapter (OSA)

Note to all users

About this publication xix

Consult assistive technologies

IBM and accessibility

Related HMC and SE console information

How to send your comments

xx IOCP User's Guide

This revision contains technical and editorial changes.

This revision contains technical and editorial changes.

This revision contains technical and editorial changes.

This revision contains technical and editorial changes.

This revision contains technical and editorial changes.

© Copyright IBM Corp. 2017, 2018 xxi

This revision contains technical and editorial changes.

This revision contains technical and editorial changes.

Summary of Changes for SB10-7037-10 System z

This revision contains technical and editorial changes.

xxii IOCP User's Guide

Summary of Changes for SB10-7037-09 System z

This revision contains technical and editorial changes.

Summary of Changes for SB10-7037-08 System z

This revision contains technical and editorial changes.

Summary of Changes for SB10-7037-07 System z

This revision contains technical and editorial changes.

Summary of changes xxiii

Summary of Changes for SB10-7037-06 System z

This revision contains technical and editorial changes.

Summary of Changes for SB10-7037-05