1 =================================================
2 Linux API for read access to z/VM Monitor Records
3 =================================================
7 Author: Gerald Schaefer (geraldsc@de.ibm.com)
14 This item delivers a new Linux API in the form of a misc char device that is
15 usable from user space and allows read access to the z/VM Monitor Records
16 collected by the `*MONITOR` System Service of z/VM.
21 The z/VM guest on which you want to access this API needs to be configured in
22 order to allow IUCV connections to the `*MONITOR` service, i.e. it needs the
23 IUCV `*MONITOR` statement in its user entry. If the monitor DCSS to be used is
24 restricted (likely), you also need the NAMESAVE <DCSS NAME> statement.
25 This item will use the IUCV device driver to access the z/VM services, so you
26 need a kernel with IUCV support. You also need z/VM version 4.4 or 5.1.
28 There are two options for being able to load the monitor DCSS (examples assume
29 that the monitor DCSS begins at 144 MB and ends at 152 MB). You can query the
30 location of the monitor DCSS with the Class E privileged CP command Q NSS MAP
31 (the values BEGPAG and ENDPAG are given in units of 4K pages).
33 See also "CP Command and Utility Reference" (SC24-6081-00) for more information
34 on the DEF STOR and Q NSS MAP commands, as well as "Saved Segments Planning
35 and Administration" (SC24-6116-00) for more information on DCSSes.
39 You can use the CP command DEF STOR CONFIG to define a "memory hole" in your
40 guest virtual storage around the address range of the DCSS.
42 Example: DEF STOR CONFIG 0.140M 200M.200M
44 This defines two blocks of storage, the first is 140MB in size an begins at
45 address 0MB, the second is 200MB in size and begins at address 200MB,
46 resulting in a total storage of 340MB. Note that the first block should
47 always start at 0 and be at least 64MB in size.
51 Your guest virtual storage has to end below the starting address of the DCSS
52 and you have to specify the "mem=" kernel parameter in your parmfile with a
53 value greater than the ending address of the DCSS.
59 This defines 140MB storage size for your guest, the parameter "mem=160M" is
60 added to the parmfile.
65 The char device is implemented as a kernel module named "monreader",
66 which can be loaded via the modprobe command, or it can be compiled into the
67 kernel instead. There is one optional module (or kernel) parameter, "mondcss",
68 to specify the name of the monitor DCSS. If the module is compiled into the
69 kernel, the kernel parameter "monreader.mondcss=<DCSS NAME>" can be specified
72 The default name for the DCSS is "MONDCSS" if none is specified. In case that
73 there are other users already connected to the `*MONITOR` service (e.g.
74 Performance Toolkit), the monitor DCSS is already defined and you have to use
75 the same DCSS. The CP command Q MONITOR (Class E privileged) shows the name
76 of the monitor DCSS, if already defined, and the users connected to the
78 Refer to the "z/VM Performance" book (SC24-6109-00) on how to create a monitor
79 DCSS if your z/VM doesn't have one already, you need Class E privileges to
80 define and save a DCSS.
87 modprobe monreader mondcss=MYDCSS
89 This loads the module and sets the DCSS name to "MYDCSS".
93 This API provides no interface to control the `*MONITOR` service, e.g. specify
94 which data should be collected. This can be done by the CP command MONITOR
95 (Class E privileged), see "CP Command and Utility Reference".
97 Device nodes with udev:
98 -----------------------
99 After loading the module, a char device will be created along with the device
100 node /<udev directory>/monreader.
102 Device nodes without udev:
103 --------------------------
104 If your distribution does not support udev, a device node will not be created
105 automatically and you have to create it manually after loading the module.
106 Therefore you need to know the major and minor numbers of the device. These
107 numbers can be found in /sys/class/misc/monreader/dev.
109 Typing cat /sys/class/misc/monreader/dev will give an output of the form
110 <major>:<minor>. The device node can be created via the mknod command, enter
111 mknod <name> c <major> <minor>, where <name> is the name of the device node
120 # cat /sys/class/misc/monreader/dev
122 # mknod /dev/monreader c 10 63
124 This loads the module with the default monitor DCSS (MONDCSS) and creates a
129 The following file operations are supported: open, release, read, poll.
130 There are two alternative methods for reading: either non-blocking read in
131 conjunction with polling, or blocking read without polling. IOCTLs are not
136 Reading from the device provides a 12 Byte monitor control element (MCE),
137 followed by a set of one or more contiguous monitor records (similar to the
138 output of the CMS utility MONWRITE without the 4K control blocks). The MCE
139 contains information on the type of the following record set (sample/event
140 data), the monitor domains contained within it and the start and end address
141 of the record set in the monitor DCSS. The start and end address can be used
142 to determine the size of the record set, the end address is the address of the
143 last byte of data. The start address is needed to handle "end-of-frame" records
144 correctly (domain 1, record 13), i.e. it can be used to determine the record
145 start offset relative to a 4K page (frame) boundary.
147 See "Appendix A: `*MONITOR`" in the "z/VM Performance" document for a description
148 of the monitor control element layout. The layout of the monitor records can
149 be found here (z/VM 5.1): https://www.vm.ibm.com/pubs/mon510/index.html
151 The layout of the data stream provided by the monreader device is as follows::
156 <first set of records> |
159 <last set of records> /
163 There may be more than one combination of MCE and corresponding record set
164 within one data set and the end of each data set is indicated by a successful
165 read with a return value of 0 (0 byte read).
166 Any received data must be considered invalid until a complete set was
167 read successfully, including the closing 0 byte read. Therefore you should
168 always read the complete set into a buffer before processing the data.
170 The maximum size of a data set can be as large as the size of the
171 monitor DCSS, so design the buffer adequately or use dynamic memory allocation.
172 The size of the monitor DCSS will be printed into syslog after loading the
173 module. You can also use the (Class E privileged) CP command Q NSS MAP to
174 list all available segments and information about them.
176 As with most char devices, error conditions are indicated by returning a
177 negative value for the number of bytes read. In this case, the errno variable
178 indicates the error condition:
181 reply failed, read data is invalid and the application
182 should discard the data read since the last successful read with 0 size.
184 copy_to_user failed, read data is invalid and the application should
185 discard the data read since the last successful read with 0 size.
187 occurs on a non-blocking read if there is no data available at the
188 moment. There is no data missing or corrupted, just try again or rather
189 use polling for non-blocking reads.
191 message limit reached, the data read since the last successful
192 read with 0 size is valid but subsequent records may be missing.
194 In the last case (EOVERFLOW) there may be missing data, in the first two cases
195 (EIO, EFAULT) there will be missing data. It's up to the application if it will
196 continue reading subsequent data or rather exit.
200 Only one user is allowed to open the char device. If it is already in use, the
201 open function will fail (return a negative value) and set errno to EBUSY.
202 The open function may also fail if an IUCV connection to the `*MONITOR` service
203 cannot be established. In this case errno will be set to EIO and an error
204 message with an IPUSER SEVER code will be printed into syslog. The IPUSER SEVER
205 codes are described in the "z/VM Performance" book, Appendix A.
209 As soon as the device is opened, incoming messages will be accepted and they
210 will account for the message limit, i.e. opening the device without reading
211 from it will provoke the "message limit reached" error (EOVERFLOW error code)