"Fossies" - the Fresh Open Source Software Archive

Member "stress-ng-0.09.56/example-jobs/cpu.job" (15 Mar 2019, 11703 Bytes) of package /linux/privat/stress-ng-0.09.56.tar.xz:


As a special service "Fossies" has tried to format the requested text file into HTML format (style: standard) with prefixed line numbers. Alternatively you can here view or download the uninterpreted source code file.

    1 #
    2 # cpu class stressors:
    3 #   various options have been commented out, one can remove the
    4 #   proceeding comment to enable these options if required.
    5 
    6 #
    7 # run the following tests in parallel or sequentially
    8 #
    9 run sequential
   10 # run parallel
   11 
   12 #
   13 # aggressive:
   14 #   enables more file, cache and memory aggressive options. This may
   15 #   slow tests down, increase latencies and  reduce  the  number  of
   16 #   bogo  ops as well as changing the balance of user time vs system
   17 #   time used depending on the type of stressor being used.
   18 #
   19 # aggressive
   20 
   21 #
   22 # ignite-cpu:
   23 #   alter kernel controls to try and maximize the CPU. This requires
   24 #   root  privilege  to alter various /sys interface controls.  Cur‐
   25 #   rently this only works for Intel P-State enabled x86 systems  on
   26 #   Linux.
   27 #
   28 # ignite-cpu
   29 
   30 #
   31 # keep-name:
   32 #   by  default,  stress-ng  will  attempt to change the name of the
   33 #   stress processes according to their functionality;  this  option
   34 #   disables  this and keeps the process names to be the name of the
   35 #   parent process, that is, stress-ng.
   36 #
   37 # keep-name
   38 
   39 #
   40 # metrics-brief:
   41 #   enable metrics and only output metrics that are non-zero.
   42 #
   43 metrics-brief
   44 
   45 #
   46 # verbose
   47 #   show all debug, warnings and normal information output.
   48 #
   49 verbose
   50 
   51 #
   52 # run each of the tests for 60 seconds
   53 #  stop stress test after N seconds. One can also specify the units
   54 #  of time in seconds, minutes, hours, days or years with the  suf‐
   55 #  fix s, m, h, d or y.
   56 #
   57 timeout 60s
   58 
   59 #
   60 # per stressor options start here
   61 #
   62 
   63 #
   64 # af-alg stressor options:
   65 #   start  N workers that exercise the AF_ALG socket domain by hash‐
   66 #   ing and encrypting various sized random messages. This exercises
   67 #   the  SHA1,  SHA224,  SHA256,  SHA384,  SHA512, MD4, MD5, RMD128,
   68 #   RMD160, RMD256, RMD320, WP256,  WP384,  WP512,  TGR128,  TGR160,
   69 #   TGR192    hashes   and   the   cbc(aes),   lrw(aes),   ofb(aes),
   70 #   xts(twofish),    xts(serpent),    xts(cast6),     xts(camellia),
   71 #
   72 af-alg 0		# 0 means 1 stressor per CPU
   73 # af-alg-ops 1000000	# stop after 1000000 bogo ops
   74 
   75 #
   76 # atomic stressor options:
   77 #   start  N workers that exercise various GCC __atomic_*() built in
   78 #   operations on 8, 16, 32 and 64 bit  intergers  that  are  shared
   79 #   among  the N workers. This stressor is only available for builds
   80 #   using GCC 4.7.4 or higher. The stressor forces  many  front  end
   81 #   cache stalls and cache references.
   82 #
   83 atomic 0		# 0 means 1 stressor per CPU
   84 # atomic-ops 1000000	# stop after 1000000 bogo ops
   85 
   86 #
   87 # bsearch stressor options:
   88 #   start  N  workers  that  binary  search a sorted array of 32 bit
   89 #   integers using bsearch(3). By default, there are 65536  elements
   90 #   in the array.  This is a useful method to exercise random access
   91 #   of memory and processor cache.
   92 #
   93 bsearch 0		# 0 means 1 stressor per CPU
   94 # bsearch-ops 1000000	# stop after 1000000 bogo ops
   95 # bsearch-size 65536	# number of elements to bsearch on
   96 
   97 #
   98 # context stressor options:
   99 #   start N workers that run three threads that  use  swapcontext(3)
  100 #   to  implement the thread-to-thread context switching. This exer‐
  101 #   cises rapid process context saving and restoring  and  is  band‐
  102 #   width limited by register and memory save and restore rates.
  103 #
  104 context 0		# 0 means 1 stressor per CPU
  105 # context-ops 1000000	# stop after 1000000 bogo ops
  106 
  107 #
  108 # cpu stressor options:
  109 #   start N workers  exercising  the  CPU  by  sequentially  working
  110 #   through  all  the different CPU stress methods. Instead of exer‐
  111 #   cising all the CPU stress methods, one can  specify  a  specific
  112 #   CPU stress method with the --cpu-method option.
  113 #
  114 cpu 0			# 0 means 1 stressor per CPU
  115 # cpu-ops 1000000	# stop after 1000000 bogo ops
  116 # cpu-load 90%		# load CPUs at 90%
  117 # cpu-load-slice 50	# 50ms per slice
  118 # cpu-method all	# use all cpu stressor methods
  119 
  120 #
  121 # cpu-online stressor options:
  122 #   start  N  workers  that  put  randomly selected CPUs offline and
  123 #   online. This Linux only stressor requires root privilege to per‐
  124 #   form this action. This is disabled by default because it can
  125 #   break older kernels.
  126 #
  127 # cpu-online 0		# 0 means 1 stressor per CPU
  128 # cpu-online-ops 1000000 # stop after 1000000 bogo ops
  129 #
  130 
  131 #
  132 # crypt stressor options:
  133 #   start  N  workers  that  encrypt  a 16 character random password
  134 #   using crypt(3).  The password is encrypted  using  MD5,  SHA-256
  135 #   and SHA-512 encryption methods.
  136 #
  137 crypt 0			# 0 means 1 stressor per CPU
  138 # crypt-ops 1000000	# stop after 1000000 bogo ops
  139 
  140 #
  141 # fp-error stressor options:
  142 #   start N workers that generate floating point exceptions.  Compu‐
  143 #   tations  are  performed to force and check for the FE_DIVBYZERO,
  144 #   FE_INEXACT, FE_INVALID, FE_OVERFLOW and FE_UNDERFLOW exceptions.
  145 #   EDOM and ERANGE errors are also checked.
  146 #
  147 fp-error 0		# 0 means 1 stressor per CPU
  148 # fp-error-ops 1000000	# stop after 1000000 bogo ops
  149 
  150 #
  151 # getrandom stressor options:
  152 #   start N workers that get 8192 random bytes from the /dev/urandom
  153 #   pool using the getrandom(2) system call (Linux) or getentropy(2)
  154 #   (OpenBSD).
  155 #
  156 getrandom 0		# 0 means 1 stressor per CPU
  157 # getrandom-ops 1000000	# stop after 1000000 bogo ops
  158 
  159 #
  160 # heapsort stressor options:
  161 #   start  N  workers  that sort 32 bit integers using the BSD heap‐
  162 #   sort.
  163 #
  164 heapsort 0		# 0 means 1 stressor per CPU
  165 # heapsort-ops 1000000	# stop after 1000000 bogo ops
  166 
  167 #
  168 # hsearch stressor options:
  169 #   start  N  workers  that  search  a  80%  full  hash  table using
  170 #   hsearch(3). By default, there are 8192  elements  inserted  into
  171 #   the  hash  table.  This is a useful method to exercise access of
  172 #   memory and processor cache.
  173 #
  174 hsearch 0		# 0 means 1 stressor per CPU
  175 # hsearch-ops 1000000	# stop after 1000000 bogo ops
  176 # hsearch-size 8192	# number of hash entries to search on
  177 
  178 #
  179 # longjmp stressor options:
  180 #   start  N  workers  that  exercise  setjmp(3)/longjmp(3) by rapid
  181 #   looping on longjmp calls.
  182 #
  183 longjmp 0		# 0 means 1 stressor per CPU
  184 # longjmp-ops 1000000	# stop after 1000000 bogo ops
  185 
  186 #
  187 # lsearch stressor options:
  188 #   start N workers that linear search a unsorted array  of  32  bit
  189 #   integers  using  lsearch(3). By default, there are 8192 elements
  190 #   in the array.  This is a useful method  to  exercise  sequential
  191 #   access of memory and processor cache.
  192 #
  193 lsearch 0		# 0 means 1 stressor per CPU
  194 # lsearch-ops 1000000	# stop after 1000000 bogo ops
  195 # lsearch-size 8192	# number of elements to lsearch on
  196 
  197 #
  198 # matrix stressor options:
  199 #   start N workers that perform various matrix operations on float‐
  200 #   ing  point values. By default, this will exercise all the matrix
  201 #   stress methods one by one.  One can specify  a  specific  matrix
  202 #   stress method with the --matrix-method option.
  203 #
  204 matrix 0		# 0 means 1 stressor per CPU
  205 # matrix-ops 1000000	# stop after 1000000 bogo ops
  206 # matrix-method all	# use all matrix stressor methods
  207 # matrix size 65536	# size of N x N matrix
  208 
  209 #
  210 # mergesort stressor options:
  211 #   start  N  workers that sort 32 bit integers using the BSD merge‐
  212 #   sort.
  213 #
  214 mergesort 0		# 0 means 1 stressor per CPU
  215 # mergesort-ops 1000000 # stop after 1000000 bogo ops
  216 # mergesort-size 262144 # number of elements to mergesort
  217 #
  218 
  219 #
  220 # nop stressor options:
  221 #   start  N  workers that consume cpu cycles issuing no-op instruc‐
  222 #   tions. This stressor is available if the assembler supports  the
  223 #   "nop" instruction.
  224 #
  225 nop 0			# 0 means 1 stressor per CPU
  226 # nop-ops 1000000	# stop after 1000000 bogo ops
  227 
  228 #
  229 # numa stressor options:
  230 #   start N workers that migrate stressors and a 4MB  memory  mapped
  231 #   buffer   around   all  the  available  NUMA  nodes.   This  uses
  232 #   migrate_pages(2)  to  move  the  stressors  and   mbind(2)   and
  233 #   move_pages(2) to move the pages of the mapped buffer. After each
  234 #   move, the buffer is written to force activity over the bus which
  235 #   results  cache misses.  This test will only run on hardware with
  236 #   NUMA enabled and more than 1 NUMA node.
  237 #
  238 numa 0			# 0 means 1 stressor per CPU
  239 # numa-ops 1000000	# stop after 1000000 bogo ops
  240 
  241 #
  242 # opcode stressor options:
  243 #   start N workers that fork off  children  that  execute  randomly
  244 #   generated  executable  code.   This will generate issues such as
  245 #   illegal instructions, bus errors,  segmentation  faults,  traps,
  246 #   floating  point errors that are handled gracefully by the stres‐
  247 #   sor.
  248 #
  249 opcode 0		# 0 means 1 stressor per CPU
  250 # opcode-ops 1000000	# stop after 1000000 bogo ops
  251 
  252 #
  253 # qsort stressor options:
  254 #   start N workers that sort 32 bit integers using qsort.
  255 #
  256 qsort 0			# 0 means 1 stressor per CPU
  257 # qsort-ops 1000000	# stop after 1000000 bogo ops
  258 # qsort-size 262144	# number of elements to qsort
  259 
  260 #
  261 # rdrand stressor options:
  262 #   start  N workers that read the Intel hardware random number gen‐
  263 #   erator (Intel Ivybridge processors upwards).
  264 #
  265 rdrand 0		# 0 means 1 stressor per CPU
  266 # rdrand-ops 1000000	# stop after 1000000 bogo ops
  267 
  268 #
  269 # str stressor options:
  270 #   start N workers that exercise various libc string  functions  on
  271 #   random strings.
  272 #
  273 str 0			# 0 means 1 stressor per CPU
  274 # str-ops 1000000	# stop after 1000000 bogo ops
  275 # str-method all	# select all string methods
  276 
  277 #
  278 # stream stressor options:
  279 #   start N workers exercising a memory bandwidth  stressor  loosely
  280 #   based  on  the STREAM "Sustainable Memory Bandwidth in High Per‐
  281 #   formance Computers" benchmarking tool by John D. McCalpin, Ph.D.
  282 #   This  stressor  allocates  buffers that are at least 4 times the
  283 #   size  of the  CPU L2 cache and  continually  performs  rounds of
  284 #   computations on large  arrays of double precision floating point
  285 #   numbers.
  286 #
  287 stream 0		# 0 means 1 stressor per CPU
  288 # stream-ops 1000000	# stop after 1000000 bogo ops
  289 # stream-l3-size 4M	# CPU level 3 cache size
  290 
  291 #
  292 # tsc stressor options:
  293 #   start N workers that read the Time Stamp Counter (TSC) 256 times
  294 #   per loop iteration (bogo operation). Available only on Intel x86
  295 #   platforms.
  296 tsc 0			# 0 means 1 stressor per CPU
  297 # tsc-ops 1000000	# stop after 1000000 bogo ops
  298 
  299 #
  300 # tsearch stressor options:
  301 #   start N workers that insert, search and delete 32  bit  integers
  302 #   on  a  binary tree using tsearch(3), tfind(3) and tdelete(3). By
  303 #   default, there are 65536 randomized integers used in  the  tree.
  304 #   This  is a useful method to exercise random access of memory and
  305 #   processor cache.
  306 #
  307 tsearch 0		# 0 means 1 stressor per CPU
  308 # tsearch-ops 1000000	# stop after 1000000 bogo ops
  309 # tsearch-size 65536	# number of elements to tsearch on
  310 
  311 #
  312 # vecmath stressor options:
  313 #   start N workers that perform various unsigned integer math oper‐
  314 #   ations on various 128 bit vectors. A mix of vector  math  opera‐
  315 #   tions  are  performed on the following vectors: 16 × 8 bits, 8 ×
  316 #   16 bits, 4 × 32 bits, 2 × 64 bits. The metrics produced by  this
  317 #   mix  depend  on  the  processor architecture and the vector math
  318 #   optimisations produced by the compiler.
  319 vecmath 0		# 0 means 1 stressor per CPU
  320 # vecmath-ops 1000000	# stop after 1000000 bogo ops
  321 
  322 #
  323 # wcs stressor options:
  324 #   start N workers that exercise various libc wide character string
  325 #   functions on random strings.
  326 #
  327 wcs 0			# 0 means 1 stressor per CPU
  328 # wcs-ops 1000000	# stop after 1000000 bogo ops
  329 # wcs-method all	# use all wide string functions
  330 
  331 #
  332 # zlib stressor option:
  333 #   start N workers compressing and decompressing random data  using
  334 #   zlib.  Each worker has two processes, one that compresses random
  335 #   data and pipes it to another process that decompresses the data.
  336 #   This stressor exercises CPU, cache and memory.
  337 #
  338 zlib 0			# 0 means 1 stressor per CPU
  339 # zlib-ops 1000000	# stop after 1000000 bogo ops
  340 # zlib-method random	# randomly select data generation methods
  341