1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387
|
#!/usr/bin/env python3
# SPDX-License-Identifier: GPL-2.0-only
#
# Copyright (c) 2020 Western Digital Corporation or its affiliates.
#
"""
# latency_percentiles.py
#
# Test the code that produces latency percentiles
# This is mostly to test the code changes to allow reporting
# of slat, clat, and lat percentiles
#
# USAGE
# python3 latency-tests.py [-f fio-path] [-a artifact-root] [--debug]
#
#
# Test scenarios:
#
# - DONE json
# unified rw reporting
# compare with latency log
# try various combinations of the ?lat_percentile options
# null, aio
# r, w, t
# - DONE json+
# check presence of latency bins
# if the json percentiles match those from the raw data
# then the latency bin values and counts are probably ok
# - DONE terse
# produce both terse, JSON output and confirm that they match
# lat only; both lat and clat
# - DONE sync_lat
# confirm that sync_lat data appears
# - MANUAL TESTING normal output:
# null ioengine
# enable all, but only clat and lat appear
# enable subset of latency types
# read, write, trim, unified
# libaio ioengine
# enable all latency types
# enable subset of latency types
# read, write, trim, unified
# ./fio/fio --name=test --randrepeat=0 --norandommap --time_based --runtime=2s --size=512M \
# --ioengine=null --slat_percentiles=1 --clat_percentiles=1 --lat_percentiles=1
# echo confirm that clat and lat percentiles appear
# ./fio/fio --name=test --randrepeat=0 --norandommap --time_based --runtime=2s --size=512M \
# --ioengine=null --slat_percentiles=0 --clat_percentiles=0 --lat_percentiles=1
# echo confirm that only lat percentiles appear
# ./fio/fio --name=test --randrepeat=0 --norandommap --time_based --runtime=2s --size=512M \
# --ioengine=null --slat_percentiles=0 --clat_percentiles=1 --lat_percentiles=0
# echo confirm that only clat percentiles appear
# ./fio/fio --name=test --randrepeat=0 --norandommap --time_based --runtime=2s --size=512M \
# --ioengine=libaio --slat_percentiles=1 --clat_percentiles=1 --lat_percentiles=1
# echo confirm that slat, clat, lat percentiles appear
# ./fio/fio --name=test --randrepeat=0 --norandommap --time_based --runtime=2s --size=512M \
# --ioengine=libaio --slat_percentiles=0 --clat_percentiles=1 --lat_percentiles=1
# echo confirm that clat and lat percentiles appear
# ./fio/fio --name=test --randrepeat=0 --norandommap --time_based --runtime=2s --size=512M \
# --ioengine=libaio -rw=randrw
# echo confirm that clat percentiles appear for reads and writes
# ./fio/fio --name=test --randrepeat=0 --norandommap --time_based --runtime=2s --size=512M \
# --ioengine=libaio --slat_percentiles=1 --clat_percentiles=0 --lat_percentiles=0 --rw=randrw
# echo confirm that slat percentiles appear for both reads and writes
# ./fio/fio --name=test --randrepeat=0 --norandommap --time_based --runtime=2s --size=512M \
# --ioengine=libaio --slat_percentiles=1 --clat_percentiles=1 --lat_percentiles=1 \
# --rw=randrw --unified_rw_reporting=1
# echo confirm that slat, clat, and lat percentiles appear for 'mixed' IOs
#./fio/fio --name=test --randrepeat=0 --norandommap --time_based --runtime=2s --size=512M \
# --ioengine=null --slat_percentiles=1 --clat_percentiles=1 --lat_percentiles=1 \
# --rw=randrw --fsync=32
# echo confirm that fsync latencies appear
"""
import os
import csv
import sys
import json
import math
import time
import argparse
import platform
import subprocess
from collections import Counter
from pathlib import Path
class FioLatTest():
"""fio latency percentile test."""
def __init__(self, artifact_root, test_options, debug):
"""
artifact_root root directory for artifacts (subdirectory will be created under here)
test test specification
"""
self.artifact_root = artifact_root
self.test_options = test_options
self.debug = debug
self.filename = None
self.json_data = None
self.terse_data = None
self.test_dir = os.path.join(self.artifact_root,
"{:03d}".format(self.test_options['test_id']))
if not os.path.exists(self.test_dir):
os.mkdir(self.test_dir)
self.filename = "latency{:03d}".format(self.test_options['test_id'])
def run_fio(self, fio_path):
"""Run a test."""
fio_args = [
"--max-jobs=16",
"--name=latency",
"--randrepeat=0",
"--norandommap",
"--time_based",
"--size=16M",
"--rwmixread=50",
"--group_reporting=1",
"--write_lat_log={0}".format(self.filename),
"--output={0}.out".format(self.filename),
"--ioengine={ioengine}".format(**self.test_options),
"--rw={rw}".format(**self.test_options),
"--runtime={runtime}".format(**self.test_options),
"--output-format={output-format}".format(**self.test_options),
]
for opt in ['slat_percentiles', 'clat_percentiles', 'lat_percentiles',
'unified_rw_reporting', 'fsync', 'fdatasync', 'numjobs',
'cmdprio_percentage', 'bssplit', 'cmdprio_bssplit']:
if opt in self.test_options:
option = '--{0}={{{0}}}'.format(opt)
fio_args.append(option.format(**self.test_options))
command = [fio_path] + fio_args
with open(os.path.join(self.test_dir, "{0}.command".format(self.filename)), "w+") as \
command_file:
command_file.write("%s\n" % command)
passed = True
stdout_file = open(os.path.join(self.test_dir, "{0}.stdout".format(self.filename)), "w+")
stderr_file = open(os.path.join(self.test_dir, "{0}.stderr".format(self.filename)), "w+")
exitcode_file = open(os.path.join(self.test_dir,
"{0}.exitcode".format(self.filename)), "w+")
try:
proc = None
# Avoid using subprocess.run() here because when a timeout occurs,
# fio will be stopped with SIGKILL. This does not give fio a
# chance to clean up and means that child processes may continue
# running and submitting IO.
proc = subprocess.Popen(command,
stdout=stdout_file,
stderr=stderr_file,
cwd=self.test_dir,
universal_newlines=True)
proc.communicate(timeout=300)
exitcode_file.write('{0}\n'.format(proc.returncode))
passed &= (proc.returncode == 0)
except subprocess.TimeoutExpired:
proc.terminate()
proc.communicate()
assert proc.poll()
print("Timeout expired")
passed = False
except Exception:
if proc:
if not proc.poll():
proc.terminate()
proc.communicate()
print("Exception: %s" % sys.exc_info())
passed = False
finally:
stdout_file.close()
stderr_file.close()
exitcode_file.close()
if passed:
if 'json' in self.test_options['output-format']:
if not self.get_json():
print('Unable to decode JSON data')
passed = False
if 'terse' in self.test_options['output-format']:
if not self.get_terse():
print('Unable to decode terse data')
passed = False
return passed
def get_json(self):
"""Convert fio JSON output into a python JSON object"""
filename = os.path.join(self.test_dir, "{0}.out".format(self.filename))
with open(filename, 'r') as file:
file_data = file.read()
#
# Sometimes fio informational messages are included at the top of the
# JSON output, especially under Windows. Try to decode output as JSON
# data, lopping off up to the first four lines
#
lines = file_data.splitlines()
for i in range(5):
file_data = '\n'.join(lines[i:])
try:
self.json_data = json.loads(file_data)
except json.JSONDecodeError:
continue
else:
return True
return False
def get_terse(self):
"""Read fio output and return terse format data."""
filename = os.path.join(self.test_dir, "{0}.out".format(self.filename))
with open(filename, 'r') as file:
file_data = file.read()
#
# Read the first few lines and see if any of them begin with '3;'
# If so, the line is probably terse output. Obviously, this only
# works for fio terse version 3 and it does not work for
# multi-line terse output
#
lines = file_data.splitlines()
for i in range(8):
file_data = lines[i]
if file_data.startswith('3;'):
self.terse_data = file_data.split(';')
return True
return False
def check_latencies(self, jsondata, ddir, slat=True, clat=True, tlat=True, plus=False,
unified=False):
"""Check fio latency data.
ddir data direction to check (0=read, 1=write, 2=trim)
slat True if submission latency data available to check
clat True if completion latency data available to check
tlat True of total latency data available to check
plus True if we actually have json+ format data where additional checks can
be carried out
unified True if fio is reporting unified r/w data
"""
types = {
'slat': slat,
'clat': clat,
'lat': tlat
}
retval = True
for lat in ['slat', 'clat', 'lat']:
this_iter = True
if not types[lat]:
if 'percentile' in jsondata[lat+'_ns']:
this_iter = False
print('unexpected %s percentiles found' % lat)
else:
print("%s percentiles skipped" % lat)
continue
else:
if 'percentile' not in jsondata[lat+'_ns']:
this_iter = False
print('%s percentiles not found in fio output' % lat)
#
# Check only for the presence/absence of json+
# latency bins. Future work can check the
# accuracy of the bin values and counts.
#
# Because the latency percentiles are based on
# the bins, we can be confident that the bin
# values and counts are correct if fio's
# latency percentiles match what we compute
# from the raw data.
#
if plus:
if 'bins' not in jsondata[lat+'_ns']:
print('bins not found with json+ output format')
this_iter = False
else:
if not self.check_jsonplus(jsondata[lat+'_ns']):
this_iter = False
else:
if 'bins' in jsondata[lat+'_ns']:
print('json+ bins found with json output format')
this_iter = False
latencies = []
for i in range(10):
lat_file = os.path.join(self.test_dir, "%s_%s.%s.log" % (self.filename, lat, i+1))
if not os.path.exists(lat_file):
break
with open(lat_file, 'r', newline='') as file:
reader = csv.reader(file)
for line in reader:
if unified or int(line[2]) == ddir:
latencies.append(int(line[1]))
if int(jsondata['total_ios']) != len(latencies):
this_iter = False
print('%s: total_ios = %s, latencies logged = %d' % \
(lat, jsondata['total_ios'], len(latencies)))
elif self.debug:
print("total_ios %s match latencies logged" % jsondata['total_ios'])
latencies.sort()
ptiles = jsondata[lat+'_ns']['percentile']
for percentile in ptiles.keys():
#
# numpy.percentile(latencies, float(percentile),
# interpolation='higher')
# produces values that mostly match what fio reports
# however, in the tails of the distribution, the values produced
# by fio's and numpy.percentile's algorithms are occasionally off
# by one latency measurement. So instead of relying on the canned
# numpy.percentile routine, implement here fio's algorithm
#
rank = math.ceil(float(percentile)/100 * len(latencies))
if rank > 0:
index = rank - 1
else:
index = 0
value = latencies[int(index)]
fio_val = int(ptiles[percentile])
# The theory in stat.h says that the proportional error will be
# less than 1/128
if not self.similar(fio_val, value):
delta = abs(fio_val - value) / value
print("Error with %s %sth percentile: "
"fio: %d, expected: %d, proportional delta: %f" %
(lat, percentile, fio_val, value, delta))
print("Rank: %d, index: %d" % (rank, index))
this_iter = False
elif self.debug:
print('%s %sth percentile values match: %d, %d' %
(lat, percentile, fio_val, value))
if this_iter:
print("%s percentiles match" % lat)
else:
retval = False
return retval
@staticmethod
def check_empty(job):
"""
Make sure JSON data is empty.
Some data structures should be empty. This function makes sure that they are.
job JSON object that we need to check for emptiness
"""
return job['total_ios'] == 0 and \
job['slat_ns']['N'] == 0 and \
job['clat_ns']['N'] == 0 and \
job['lat_ns']['N'] == 0
def check_nocmdprio_lat(self, job):
"""
Make sure no per priority latencies appear.
job JSON object to check
"""
for ddir in ['read', 'write', 'trim']:
if ddir in job:
if 'prios' in job[ddir]:
print("Unexpected per priority latencies found in %s output" % ddir)
return False
if self.debug:
print("No per priority latencies found")
return True
@staticmethod
def similar(approximation, actual):
"""
Check whether the approximate values recorded by fio are within the theoretical bound.
Since it is impractical to store exact latency measurements for each and every IO, fio
groups similar latency measurements into variable-sized bins. The theory in stat.h says
that the proportional error will be less than 1/128. This function checks whether this
is true.
TODO This test will fail when comparing a value from the largest latency bin against its
actual measurement. Find some way to detect this and avoid failing.
approximation value of the bin used by fio to store a given latency
actual actual latency value
"""
# Avoid a division by zero. The smallest latency values have no error.
if actual == 0:
return approximation == 0
delta = abs(approximation - actual) / actual
return delta <= 1/128
def check_jsonplus(self, jsondata):
"""Check consistency of json+ data
When we have json+ data we can check the min value, max value, and
sample size reported by fio
jsondata json+ data that we need to check
"""
retval = True
keys = [int(k) for k in jsondata['bins'].keys()]
values = [int(jsondata['bins'][k]) for k in jsondata['bins'].keys()]
smallest = min(keys)
biggest = max(keys)
sampsize = sum(values)
if not self.similar(jsondata['min'], smallest):
retval = False
print('reported min %d does not match json+ min %d' % (jsondata['min'], smallest))
elif self.debug:
print('json+ min values match: %d' % jsondata['min'])
if not self.similar(jsondata['max'], biggest):
retval = False
print('reported max %d does not match json+ max %d' % (jsondata['max'], biggest))
elif self.debug:
print('json+ max values match: %d' % jsondata['max'])
if sampsize != jsondata['N']:
retval = False
print('reported sample size %d does not match json+ total count %d' % \
(jsondata['N'], sampsize))
elif self.debug:
print('json+ sample sizes match: %d' % sampsize)
return retval
def check_sync_lat(self, jsondata, plus=False):
"""Check fsync latency percentile data.
All we can check is that some percentiles are reported, unless we have json+ data.
If we actually have json+ data then we can do more checking.
jsondata JSON data for fsync operations
plus True if we actually have json+ data
"""
retval = True
if 'percentile' not in jsondata['lat_ns']:
print("Sync percentile data not found")
return False
if int(jsondata['total_ios']) != int(jsondata['lat_ns']['N']):
retval = False
print('Mismatch between total_ios and lat_ns sample size')
elif self.debug:
print('sync sample sizes match: %d' % jsondata['total_ios'])
if not plus:
if 'bins' in jsondata['lat_ns']:
print('Unexpected json+ bin data found')
return False
if not self.check_jsonplus(jsondata['lat_ns']):
retval = False
return retval
def check_terse(self, terse, jsondata):
"""Compare terse latencies with JSON latencies.
terse terse format data for checking
jsondata JSON format data for checking
"""
retval = True
for lat in terse:
split = lat.split('%')
pct = split[0]
terse_val = int(split[1][1:])
json_val = math.floor(jsondata[pct]/1000)
if terse_val != json_val:
retval = False
print('Mismatch with %sth percentile: json value=%d,%d terse value=%d' % \
(pct, jsondata[pct], json_val, terse_val))
elif self.debug:
print('Terse %sth percentile matches JSON value: %d' % (pct, terse_val))
return retval
def check_prio_latencies(self, jsondata, clat=True, plus=False):
"""Check consistency of per priority latencies.
clat True if we should check clat data; other check lat data
plus True if we have json+ format data where additional checks can
be carried out
unified True if fio is reporting unified r/w data
"""
if clat:
obj = combined = 'clat_ns'
else:
obj = combined = 'lat_ns'
if not 'prios' in jsondata or not combined in jsondata:
print("Error identifying per priority latencies")
return False
sum_sample_size = sum([x[obj]['N'] for x in jsondata['prios']])
if sum_sample_size != jsondata[combined]['N']:
print("Per prio sample size sum %d != combined sample size %d" %
(sum_sample_size, jsondata[combined]['N']))
return False
elif self.debug:
print("Per prio sample size sum %d == combined sample size %d" %
(sum_sample_size, jsondata[combined]['N']))
min_val = min([x[obj]['min'] for x in jsondata['prios']])
if min_val != jsondata[combined]['min']:
print("Min per prio min latency %d does not match min %d from combined data" %
(min_val, jsondata[combined]['min']))
return False
elif self.debug:
print("Min per prio min latency %d matches min %d from combined data" %
(min_val, jsondata[combined]['min']))
max_val = max([x[obj]['max'] for x in jsondata['prios']])
if max_val != jsondata[combined]['max']:
print("Max per prio max latency %d does not match max %d from combined data" %
(max_val, jsondata[combined]['max']))
return False
elif self.debug:
print("Max per prio max latency %d matches max %d from combined data" %
(max_val, jsondata[combined]['max']))
weighted_vals = [x[obj]['mean'] * x[obj]['N'] for x in jsondata['prios']]
weighted_avg = sum(weighted_vals) / jsondata[combined]['N']
delta = abs(weighted_avg - jsondata[combined]['mean'])
if (delta / jsondata[combined]['mean']) > 0.0001:
print("Difference between merged per prio weighted average %f mean "
"and actual mean %f exceeds 0.01%%" % (weighted_avg, jsondata[combined]['mean']))
return False
elif self.debug:
print("Merged per prio weighted average %f mean matches actual mean %f" %
(weighted_avg, jsondata[combined]['mean']))
if plus:
for prio in jsondata['prios']:
if not self.check_jsonplus(prio[obj]):
return False
counter = Counter()
for prio in jsondata['prios']:
counter.update(prio[obj]['bins'])
bins = dict(counter)
if len(bins) != len(jsondata[combined]['bins']):
print("Number of merged bins %d does not match number of overall bins %d" %
(len(bins), len(jsondata[combined]['bins'])))
return False
elif self.debug:
print("Number of merged bins %d matches number of overall bins %d" %
(len(bins), len(jsondata[combined]['bins'])))
for duration in bins.keys():
if bins[duration] != jsondata[combined]['bins'][duration]:
print("Merged per prio count does not match overall count for duration %d" %
duration)
return False
print("Merged per priority latency data match combined latency data")
return True
def check(self):
"""Check test output."""
raise NotImplementedError()
class Test001(FioLatTest):
"""Test object for Test 1."""
def check(self):
"""Check Test 1 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['write']):
print("Unexpected write data found in output")
retval = False
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['read'], 0, slat=False)
return retval
class Test002(FioLatTest):
"""Test object for Test 2."""
def check(self):
"""Check Test 2 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['read']):
print("Unexpected read data found in output")
retval = False
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['write'], 1, slat=False, clat=False)
return retval
class Test003(FioLatTest):
"""Test object for Test 3."""
def check(self):
"""Check Test 3 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['read']):
print("Unexpected read data found in output")
retval = False
if not self.check_empty(job['write']):
print("Unexpected write data found in output")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['trim'], 2, slat=False, tlat=False)
return retval
class Test004(FioLatTest):
"""Test object for Tests 4, 13."""
def check(self):
"""Check Test 4, 13 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['write']):
print("Unexpected write data found in output")
retval = False
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['read'], 0, plus=True)
return retval
class Test005(FioLatTest):
"""Test object for Test 5."""
def check(self):
"""Check Test 5 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['read']):
print("Unexpected read data found in output")
retval = False
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['write'], 1, slat=False, plus=True)
return retval
class Test006(FioLatTest):
"""Test object for Test 6."""
def check(self):
"""Check Test 6 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['write']):
print("Unexpected write data found in output")
retval = False
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['read'], 0, slat=False, tlat=False, plus=True)
return retval
class Test007(FioLatTest):
"""Test object for Test 7."""
def check(self):
"""Check Test 7 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['read'], 0, clat=False, tlat=False, plus=True)
retval &= self.check_latencies(job['write'], 1, clat=False, tlat=False, plus=True)
return retval
class Test008(FioLatTest):
"""Test object for Tests 8, 14."""
def check(self):
"""Check Test 8, 14 output."""
job = self.json_data['jobs'][0]
retval = True
if 'read' in job or 'write' in job or 'trim' in job:
print("Unexpected data direction found in fio output")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['mixed'], 0, plus=True, unified=True)
return retval
class Test009(FioLatTest):
"""Test object for Test 9."""
def check(self):
"""Check Test 9 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['read']):
print("Unexpected read data found in output")
retval = False
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
if not self.check_sync_lat(job['sync'], plus=True):
print("Error checking fsync latency data")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['write'], 1, slat=False, plus=True)
return retval
class Test010(FioLatTest):
"""Test object for Test 10."""
def check(self):
"""Check Test 10 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['read'], 0, plus=True)
retval &= self.check_latencies(job['write'], 1, plus=True)
retval &= self.check_terse(self.terse_data[17:34], job['read']['lat_ns']['percentile'])
retval &= self.check_terse(self.terse_data[58:75], job['write']['lat_ns']['percentile'])
# Terse data checking only works for default percentiles.
# This needs to be changed if something other than the default is ever used.
return retval
class Test011(FioLatTest):
"""Test object for Test 11."""
def check(self):
"""Check Test 11 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
if not self.check_nocmdprio_lat(job):
print("Unexpected per priority latencies found")
retval = False
retval &= self.check_latencies(job['read'], 0, slat=False, clat=False, plus=True)
retval &= self.check_latencies(job['write'], 1, slat=False, clat=False, plus=True)
retval &= self.check_terse(self.terse_data[17:34], job['read']['lat_ns']['percentile'])
retval &= self.check_terse(self.terse_data[58:75], job['write']['lat_ns']['percentile'])
# Terse data checking only works for default percentiles.
# This needs to be changed if something other than the default is ever used.
return retval
class Test015(FioLatTest):
"""Test object for Test 15."""
def check(self):
"""Check Test 15 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['write']):
print("Unexpected write data found in output")
retval = False
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
retval &= self.check_latencies(job['read'], 0, plus=True)
retval &= self.check_prio_latencies(job['read'], clat=False, plus=True)
return retval
class Test016(FioLatTest):
"""Test object for Test 16."""
def check(self):
"""Check Test 16 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['read']):
print("Unexpected read data found in output")
retval = False
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
retval &= self.check_latencies(job['write'], 1, slat=False, plus=True)
retval &= self.check_prio_latencies(job['write'], clat=False, plus=True)
return retval
class Test017(FioLatTest):
"""Test object for Test 17."""
def check(self):
"""Check Test 17 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['write']):
print("Unexpected write data found in output")
retval = False
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
retval &= self.check_latencies(job['read'], 0, slat=False, tlat=False, plus=True)
retval &= self.check_prio_latencies(job['read'], plus=True)
return retval
class Test018(FioLatTest):
"""Test object for Test 18."""
def check(self):
"""Check Test 18 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
retval &= self.check_latencies(job['read'], 0, clat=False, tlat=False, plus=True)
retval &= self.check_latencies(job['write'], 1, clat=False, tlat=False, plus=True)
# We actually have json+ data but setting plus=False below avoids checking the
# json+ bins which did not exist for clat and lat because this job is run with
# clat_percentiles=0, lat_percentiles=0, However, we can still check the summary
# statistics
retval &= self.check_prio_latencies(job['write'], plus=False)
retval &= self.check_prio_latencies(job['read'], plus=False)
return retval
class Test019(FioLatTest):
"""Test object for Tests 19, 20."""
def check(self):
"""Check Test 19, 20 output."""
job = self.json_data['jobs'][0]
retval = True
if 'read' in job or 'write' in job or 'trim' in job:
print("Unexpected data direction found in fio output")
retval = False
retval &= self.check_latencies(job['mixed'], 0, plus=True, unified=True)
retval &= self.check_prio_latencies(job['mixed'], clat=False, plus=True)
return retval
class Test021(FioLatTest):
"""Test object for Test 21."""
def check(self):
"""Check Test 21 output."""
job = self.json_data['jobs'][0]
retval = True
if not self.check_empty(job['trim']):
print("Unexpected trim data found in output")
retval = False
retval &= self.check_latencies(job['read'], 0, slat=False, tlat=False, plus=True)
retval &= self.check_latencies(job['write'], 1, slat=False, tlat=False, plus=True)
retval &= self.check_prio_latencies(job['read'], clat=True, plus=True)
retval &= self.check_prio_latencies(job['write'], clat=True, plus=True)
return retval
def parse_args():
"""Parse command-line arguments."""
parser = argparse.ArgumentParser()
parser.add_argument('-f', '--fio', help='path to file executable (e.g., ./fio)')
parser.add_argument('-a', '--artifact-root', help='artifact root directory')
parser.add_argument('-d', '--debug', help='enable debug output', action='store_true')
parser.add_argument('-s', '--skip', nargs='+', type=int,
help='list of test(s) to skip')
parser.add_argument('-o', '--run-only', nargs='+', type=int,
help='list of test(s) to run, skipping all others')
args = parser.parse_args()
return args
def main():
"""Run tests of fio latency percentile reporting"""
args = parse_args()
artifact_root = args.artifact_root if args.artifact_root else \
"latency-test-{0}".format(time.strftime("%Y%m%d-%H%M%S"))
os.mkdir(artifact_root)
print("Artifact directory is %s" % artifact_root)
if args.fio:
fio = str(Path(args.fio).absolute())
else:
fio = 'fio'
print("fio path is %s" % fio)
if platform.system() == 'Linux':
aio = 'libaio'
elif platform.system() == 'Windows':
aio = 'windowsaio'
else:
aio = 'posixaio'
test_list = [
{
# randread, null
# enable slat, clat, lat
# only clat and lat will appear because
# because the null ioengine is synchronous
"test_id": 1,
"runtime": 2,
"output-format": "json",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": 'null',
'rw': 'randread',
"test_obj": Test001,
},
{
# randwrite, null
# enable lat only
"test_id": 2,
"runtime": 2,
"output-format": "json",
"slat_percentiles": 0,
"clat_percentiles": 0,
"lat_percentiles": 1,
"ioengine": 'null',
'rw': 'randwrite',
"test_obj": Test002,
},
{
# randtrim, null
# enable clat only
"test_id": 3,
"runtime": 2,
"output-format": "json",
"slat_percentiles": 0,
"clat_percentiles": 1,
"lat_percentiles": 0,
"ioengine": 'null',
'rw': 'randtrim',
"test_obj": Test003,
},
{
# randread, aio
# enable slat, clat, lat
# all will appear because libaio is asynchronous
"test_id": 4,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randread',
"test_obj": Test004,
},
{
# randwrite, aio
# enable only clat, lat
"test_id": 5,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 0,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randwrite',
"test_obj": Test005,
},
{
# randread, aio
# by default only clat should appear
"test_id": 6,
"runtime": 5,
"output-format": "json+",
"ioengine": aio,
'rw': 'randread',
"test_obj": Test006,
},
{
# 50/50 r/w, aio
# enable only slat
"test_id": 7,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 1,
"clat_percentiles": 0,
"lat_percentiles": 0,
"ioengine": aio,
'rw': 'randrw',
"test_obj": Test007,
},
{
# 50/50 r/w, aio, unified_rw_reporting
# enable slat, clat, lat
"test_id": 8,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randrw',
'unified_rw_reporting': 1,
"test_obj": Test008,
},
{
# randwrite, null
# enable slat, clat, lat
# fsync
"test_id": 9,
"runtime": 2,
"output-format": "json+",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": 'null',
'rw': 'randwrite',
'fsync': 32,
"test_obj": Test009,
},
{
# 50/50 r/w, aio
# enable slat, clat, lat
"test_id": 10,
"runtime": 5,
"output-format": "terse,json+",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randrw',
"test_obj": Test010,
},
{
# 50/50 r/w, aio
# enable only lat
"test_id": 11,
"runtime": 5,
"output-format": "terse,json+",
"slat_percentiles": 0,
"clat_percentiles": 0,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randrw',
"test_obj": Test011,
},
{
# randread, null
# enable slat, clat, lat
# only clat and lat will appear because
# because the null ioengine is synchronous
# same as Test 1 except add numjobs = 4 to test
# sum_thread_stats() changes
"test_id": 12,
"runtime": 2,
"output-format": "json",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": 'null',
'rw': 'randread',
'numjobs': 4,
"test_obj": Test001,
},
{
# randread, aio
# enable slat, clat, lat
# all will appear because libaio is asynchronous
# same as Test 4 except add numjobs = 4 to test
# sum_thread_stats() changes
"test_id": 13,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randread',
'numjobs': 4,
"test_obj": Test004,
},
{
# 50/50 r/w, aio, unified_rw_reporting
# enable slat, clat, lata
# same as Test 8 except add numjobs = 4 to test
# sum_thread_stats() changes
"test_id": 14,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randrw',
'unified_rw_reporting': 1,
'numjobs': 4,
"test_obj": Test008,
},
{
# randread, aio
# enable slat, clat, lat
# all will appear because libaio is asynchronous
# same as Test 4 except add cmdprio_percentage
"test_id": 15,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randread',
'cmdprio_percentage': 50,
"test_obj": Test015,
},
{
# randwrite, aio
# enable only clat, lat
# same as Test 5 except add cmdprio_percentage
"test_id": 16,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 0,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randwrite',
'cmdprio_percentage': 50,
"test_obj": Test016,
},
{
# randread, aio
# by default only clat should appear
# same as Test 6 except add cmdprio_percentage
"test_id": 17,
"runtime": 5,
"output-format": "json+",
"ioengine": aio,
'rw': 'randread',
'cmdprio_percentage': 50,
"test_obj": Test017,
},
{
# 50/50 r/w, aio
# enable only slat
# same as Test 7 except add cmdprio_percentage
"test_id": 18,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 1,
"clat_percentiles": 0,
"lat_percentiles": 0,
"ioengine": aio,
'rw': 'randrw',
'cmdprio_percentage': 50,
"test_obj": Test018,
},
{
# 50/50 r/w, aio, unified_rw_reporting
# enable slat, clat, lat
# same as Test 8 except add cmdprio_percentage
"test_id": 19,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randrw',
'unified_rw_reporting': 1,
'cmdprio_percentage': 50,
"test_obj": Test019,
},
{
# 50/50 r/w, aio, unified_rw_reporting
# enable slat, clat, lat
# same as Test 19 except add numjobs = 4 to test
# sum_thread_stats() changes
"test_id": 20,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 1,
"clat_percentiles": 1,
"lat_percentiles": 1,
"ioengine": aio,
'rw': 'randrw',
'unified_rw_reporting': 1,
'cmdprio_percentage': 50,
'numjobs': 4,
"test_obj": Test019,
},
{
# r/w, aio
# enable only clat
# test bssplit and cmdprio_bssplit
"test_id": 21,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 0,
"clat_percentiles": 1,
"lat_percentiles": 0,
"ioengine": aio,
'rw': 'randrw',
'bssplit': '64k/40:1024k/60',
'cmdprio_bssplit': '64k/25/1/1:64k/75/3/2:1024k/0',
"test_obj": Test021,
},
{
# r/w, aio
# enable only clat
# same as Test 21 except add numjobs = 4 to test
# sum_thread_stats() changes
"test_id": 22,
"runtime": 5,
"output-format": "json+",
"slat_percentiles": 0,
"clat_percentiles": 1,
"lat_percentiles": 0,
"ioengine": aio,
'rw': 'randrw',
'bssplit': '64k/40:1024k/60',
'cmdprio_bssplit': '64k/25/1/1:64k/75/3/2:1024k/0',
'numjobs': 4,
"test_obj": Test021,
},
]
passed = 0
failed = 0
skipped = 0
for test in test_list:
if (args.skip and test['test_id'] in args.skip) or \
(args.run_only and test['test_id'] not in args.run_only):
skipped = skipped + 1
outcome = 'SKIPPED (User request)'
elif (platform.system() != 'Linux' or os.geteuid() != 0) and \
('cmdprio_percentage' in test or 'cmdprio_bssplit' in test):
skipped = skipped + 1
outcome = 'SKIPPED (Linux root required for cmdprio tests)'
else:
test_obj = test['test_obj'](artifact_root, test, args.debug)
status = test_obj.run_fio(fio)
if status:
status = test_obj.check()
if status:
passed = passed + 1
outcome = 'PASSED'
else:
failed = failed + 1
outcome = 'FAILED'
print("**********Test {0} {1}**********".format(test['test_id'], outcome))
print("{0} tests passed, {1} failed, {2} skipped".format(passed, failed, skipped))
sys.exit(failed)
if __name__ == '__main__':
main()
|