monreader.txt 8.4 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197
  1. Date : 2004-Nov-26
  2. Author: Gerald Schaefer (geraldsc@de.ibm.com)
  3. Linux API for read access to z/VM Monitor Records
  4. =================================================
  5. Description
  6. ===========
  7. This item delivers a new Linux API in the form of a misc char device that is
  8. usable from user space and allows read access to the z/VM Monitor Records
  9. collected by the *MONITOR System Service of z/VM.
  10. User Requirements
  11. =================
  12. The z/VM guest on which you want to access this API needs to be configured in
  13. order to allow IUCV connections to the *MONITOR service, i.e. it needs the
  14. IUCV *MONITOR statement in its user entry. If the monitor DCSS to be used is
  15. restricted (likely), you also need the NAMESAVE <DCSS NAME> statement.
  16. This item will use the IUCV device driver to access the z/VM services, so you
  17. need a kernel with IUCV support. You also need z/VM version 4.4 or 5.1.
  18. There are two options for being able to load the monitor DCSS (examples assume
  19. that the monitor DCSS begins at 144 MB and ends at 152 MB). You can query the
  20. location of the monitor DCSS with the Class E privileged CP command Q NSS MAP
  21. (the values BEGPAG and ENDPAG are given in units of 4K pages).
  22. See also "CP Command and Utility Reference" (SC24-6081-00) for more information
  23. on the DEF STOR and Q NSS MAP commands, as well as "Saved Segments Planning
  24. and Administration" (SC24-6116-00) for more information on DCSSes.
  25. 1st option:
  26. -----------
  27. You can use the CP command DEF STOR CONFIG to define a "memory hole" in your
  28. guest virtual storage around the address range of the DCSS.
  29. Example: DEF STOR CONFIG 0.140M 200M.200M
  30. This defines two blocks of storage, the first is 140MB in size an begins at
  31. address 0MB, the second is 200MB in size and begins at address 200MB,
  32. resulting in a total storage of 340MB. Note that the first block should
  33. always start at 0 and be at least 64MB in size.
  34. 2nd option:
  35. -----------
  36. Your guest virtual storage has to end below the starting address of the DCSS
  37. and you have to specify the "mem=" kernel parameter in your parmfile with a
  38. value greater than the ending address of the DCSS.
  39. Example: DEF STOR 140M
  40. This defines 140MB storage size for your guest, the parameter "mem=160M" is
  41. added to the parmfile.
  42. User Interface
  43. ==============
  44. The char device is implemented as a kernel module named "monreader",
  45. which can be loaded via the modprobe command, or it can be compiled into the
  46. kernel instead. There is one optional module (or kernel) parameter, "mondcss",
  47. to specify the name of the monitor DCSS. If the module is compiled into the
  48. kernel, the kernel parameter "monreader.mondcss=<DCSS NAME>" can be specified
  49. in the parmfile.
  50. The default name for the DCSS is "MONDCSS" if none is specified. In case that
  51. there are other users already connected to the *MONITOR service (e.g.
  52. Performance Toolkit), the monitor DCSS is already defined and you have to use
  53. the same DCSS. The CP command Q MONITOR (Class E privileged) shows the name
  54. of the monitor DCSS, if already defined, and the users connected to the
  55. *MONITOR service.
  56. Refer to the "z/VM Performance" book (SC24-6109-00) on how to create a monitor
  57. DCSS if your z/VM doesn't have one already, you need Class E privileges to
  58. define and save a DCSS.
  59. Example:
  60. --------
  61. modprobe monreader mondcss=MYDCSS
  62. This loads the module and sets the DCSS name to "MYDCSS".
  63. NOTE:
  64. -----
  65. This API provides no interface to control the *MONITOR service, e.g. specify
  66. which data should be collected. This can be done by the CP command MONITOR
  67. (Class E privileged), see "CP Command and Utility Reference".
  68. Device nodes with udev:
  69. -----------------------
  70. After loading the module, a char device will be created along with the device
  71. node /<udev directory>/monreader.
  72. Device nodes without udev:
  73. --------------------------
  74. If your distribution does not support udev, a device node will not be created
  75. automatically and you have to create it manually after loading the module.
  76. Therefore you need to know the major and minor numbers of the device. These
  77. numbers can be found in /sys/class/misc/monreader/dev.
  78. Typing cat /sys/class/misc/monreader/dev will give an output of the form
  79. <major>:<minor>. The device node can be created via the mknod command, enter
  80. mknod <name> c <major> <minor>, where <name> is the name of the device node
  81. to be created.
  82. Example:
  83. --------
  84. # modprobe monreader
  85. # cat /sys/class/misc/monreader/dev
  86. 10:63
  87. # mknod /dev/monreader c 10 63
  88. This loads the module with the default monitor DCSS (MONDCSS) and creates a
  89. device node.
  90. File operations:
  91. ----------------
  92. The following file operations are supported: open, release, read, poll.
  93. There are two alternative methods for reading: either non-blocking read in
  94. conjunction with polling, or blocking read without polling. IOCTLs are not
  95. supported.
  96. Read:
  97. -----
  98. Reading from the device provides a 12 Byte monitor control element (MCE),
  99. followed by a set of one or more contiguous monitor records (similar to the
  100. output of the CMS utility MONWRITE without the 4K control blocks). The MCE
  101. contains information on the type of the following record set (sample/event
  102. data), the monitor domains contained within it and the start and end address
  103. of the record set in the monitor DCSS. The start and end address can be used
  104. to determine the size of the record set, the end address is the address of the
  105. last byte of data. The start address is needed to handle "end-of-frame" records
  106. correctly (domain 1, record 13), i.e. it can be used to determine the record
  107. start offset relative to a 4K page (frame) boundary.
  108. See "Appendix A: *MONITOR" in the "z/VM Performance" document for a description
  109. of the monitor control element layout. The layout of the monitor records can
  110. be found here (z/VM 5.1): http://www.vm.ibm.com/pubs/mon510/index.html
  111. The layout of the data stream provided by the monreader device is as follows:
  112. ...
  113. <0 byte read>
  114. <first MCE> \
  115. <first set of records> |
  116. ... |- data set
  117. <last MCE> |
  118. <last set of records> /
  119. <0 byte read>
  120. ...
  121. There may be more than one combination of MCE and corresponding record set
  122. within one data set and the end of each data set is indicated by a successful
  123. read with a return value of 0 (0 byte read).
  124. Any received data must be considered invalid until a complete set was
  125. read successfully, including the closing 0 byte read. Therefore you should
  126. always read the complete set into a buffer before processing the data.
  127. The maximum size of a data set can be as large as the size of the
  128. monitor DCSS, so design the buffer adequately or use dynamic memory allocation.
  129. The size of the monitor DCSS will be printed into syslog after loading the
  130. module. You can also use the (Class E privileged) CP command Q NSS MAP to
  131. list all available segments and information about them.
  132. As with most char devices, error conditions are indicated by returning a
  133. negative value for the number of bytes read. In this case, the errno variable
  134. indicates the error condition:
  135. EIO: reply failed, read data is invalid and the application
  136. should discard the data read since the last successful read with 0 size.
  137. EFAULT: copy_to_user failed, read data is invalid and the application should
  138. discard the data read since the last successful read with 0 size.
  139. EAGAIN: occurs on a non-blocking read if there is no data available at the
  140. moment. There is no data missing or corrupted, just try again or rather
  141. use polling for non-blocking reads.
  142. EOVERFLOW: message limit reached, the data read since the last successful
  143. read with 0 size is valid but subsequent records may be missing.
  144. In the last case (EOVERFLOW) there may be missing data, in the first two cases
  145. (EIO, EFAULT) there will be missing data. It's up to the application if it will
  146. continue reading subsequent data or rather exit.
  147. Open:
  148. -----
  149. Only one user is allowed to open the char device. If it is already in use, the
  150. open function will fail (return a negative value) and set errno to EBUSY.
  151. The open function may also fail if an IUCV connection to the *MONITOR service
  152. cannot be established. In this case errno will be set to EIO and an error
  153. message with an IPUSER SEVER code will be printed into syslog. The IPUSER SEVER
  154. codes are described in the "z/VM Performance" book, Appendix A.
  155. NOTE:
  156. -----
  157. As soon as the device is opened, incoming messages will be accepted and they
  158. will account for the message limit, i.e. opening the device without reading
  159. from it will provoke the "message limit reached" error (EOVERFLOW error code)
  160. eventually.