308:: for example, code which uses the processor but produces internal results which are not used or output) make the use and design of synthetic benchmarks more difficult. Version 2.0 of the benchmark, released by Weicker and Richardson in March 1988, had a number of changes intended to foil a range of compiler techniques. Yet it was carefully crafted so as not to change the underlying benchmark. This effort to foil compilers was only partly successful. Dhrystone 2.1, released in May of the same year, had some minor changes and as of July 2010 remains the current definition of Dhrystone.
240:
36:
340:) can confound simple comparisons. For example, the same high-level task may require many more instructions on a RISC machine, but might execute faster than a single CISC instruction. Thus, the Dhrystone score counts only the number of program iteration completions per second, allowing individual machines to perform this calculation in a machine-specific way. Another common representation of the Dhrystone benchmark is the
312:
operations, which is largely language-related: both Ada and Pascal have strings as normal variables in the language, whereas C does not, so what was simple variable assignment in reference benchmarks became buffer copy operations in the C library. Another issue is that the score reported does not include information which is critical when comparing systems such as which compiler was used, and what optimizations.
315:
Dhrystone remains remarkably resilient as a simple benchmark, but its continuing value in establishing true performance is questionable. It is easy to use, well documented, fully self-contained, well understood, and can be made to work on almost any system. In particular, it has remained in broad
374:
It is susceptible to compiler optimizations. For example, it does a lot of string copying in an attempt to measure string copying performance. However, the strings in
Dhrystone are of known constant length and their starts are aligned on natural boundaries, two characteristics usually absent from
225:
benchmarks, meaning that they are simple programs that are carefully designed to statistically mimic the processor usage of some common set of programs. Whetstone, developed in 1972, originally strove to mimic typical Algol 60 programs based on measurements from 1970, but eventually became most
311:
Other than issues related to compiler optimization, various other issues have been cited with the
Dhrystone. Most of these, including the small code size and small data set size, were understood at the time of its publication in 1984. More subtle is the slight over-representation of string
375:
real programs. Therefore, an optimizer can replace a string copy with a sequence of word moves without any loops, which will be much faster. This optimization consequently overstates system performance, sometimes by more than 30%.
324:
standalone benchmark, HINT, Stream, and even
Bytemark are widely quoted and used, as well as more specific benchmarks for the memory subsystem (Cachebench), TCP/IP (TTCP), and many others.
300:
Dhrystone's eventual importance as an indicator of general-purpose ("integer") performance of new computers made it a target for commercial compiler writers. Various modern compiler
332:
Dhrystone may represent a result more meaningfully than MIPS (million instructions per second) because instruction count comparisons between different instruction sets (e.g.
218:
benchmark for floating point operations. The output from the benchmark is the number of
Dhrystones per second (the number of iterations of the main code loop per second).
355:
Another way to represent results is in DMIPS/MHz, where DMIPS result is further divided by CPU frequency, to allow for easier comparison of CPUs running at different
394:
benchmark was created in 1988 to include a suite of (initially 8) much larger programs (including a compiler) which could not fit into L1 or L2 caches of that era.
144:
465:
583:
404:
190:, assignments, etc. From this he wrote the Dhrystone benchmark to correspond to a representative mix. Dhrystone was published in
498:
100:
72:
287:
119:
269:
337:
578:
345:
152:
79:
265:
57:
183:
348:) obtained when the Dhrystone score is divided by 1757 (the number of Dhrystones per second obtained on the
86:
170:
With
Dhrystone, Weicker gathered meta-data from a broad range of software, including programs written in
261:
226:
popular in its
Fortran version, reflecting the highly numerical orientation of computing in the 1960s.
191:
186:. He then characterized these programs in terms of various common constructs: procedure calls, pointer
53:
68:
524:
458:
250:
195:
254:
46:
390:, thus not exercising data cache miss performance. To counter fits-in-the-cache problem, the
305:
215:
160:
140:
17:
301:
386:
performance is not rigorously tested. Similarly, Dhrystone may also fit completely in the
8:
371:
It features unusual code that is not usually representative of modern real-life programs.
543:
159:) performance. The name "Dhrystone" is a pun on a different benchmark algorithm called
93:
494:
383:
379:
202:
developed by Rick
Richardson ("version 1.1") greatly contributing to its popularity.
547:
533:
433:
211:
164:
572:
556:
538:
519:
562:
187:
387:
356:
459:"Dhrystone Benchmark: History, Analysis, "Scores" and Recommendations"
410:
491:
Microcontrollers: Hardware and
Firmware for 8-bit and 32-bit devices
239:
35:
321:
316:
use in the embedded computing world, though the recently developed
179:
557:
Dhrystone
Benchmark: Rationale for Version 2 and Measurement Rules
391:
171:
148:
151:) programming. The Dhrystone grew to become representative of
317:
333:
199:
175:
429:
349:
156:
214:
operations, thus the name is a pun on the then-popular
520:"Dhrystone: A Synthetic Systems Programming Benchmark"
60:. Unsourced material may be challenged and removed.
570:
229:
367:Using Dhrystone as a benchmark has pitfalls:
488:
136:
405:Standard Performance Evaluation Corporation
378:Dhrystone's small code size may fit in the
268:. Unsourced material may be challenged and
493:. Società Editrice Esculapio. p. 66.
205:
537:
288:Learn how and when to remove this message
147:intended to be representative of system (
120:Learn how and when to remove this message
517:
452:
450:
14:
584:Computer-related introductions in 1984
571:
447:
266:adding citations to reliable sources
233:
210:The Dhrystone benchmark contains no
58:adding citations to reliable sources
29:
24:
518:Weicker, Reinhold (October 1984).
25:
595:
511:
456:
221:Both Whetstone and Dhrystone are
238:
34:
471:from the original on 2011-07-26
362:
352:, nominally a 1 MIPS machine).
45:needs additional citations for
482:
423:
13:
1:
416:
230:Issues addressed by Dhrystone
143:program developed in 1984 by
27:Computer benchmarking program
7:
565:(Reinhold P. Weicker, 1995)
563:DHRYSTONE Benchmark Program
559:(Reinhold P. Weicker, 1988)
440:→ wet stone → dry stone → d
398:
10:
600:
327:
525:Communications of the ACM
382:of a modern CPU, so that
306:elimination of dead code
206:Dhrystone vs. Whetstone
579:Benchmarks (computing)
539:10.1145/358274.358283
489:Franco Zappa (2017).
320:benchmark suite, the
304:techniques (such as
302:static code analysis
262:improve this section
54:improve this article
163:, which emphasizes
145:Reinhold P. Weicker
500:978-88-9385-022-3
384:instruction fetch
380:instruction cache
298:
297:
290:
153:general processor
130:
129:
122:
104:
16:(Redirected from
591:
551:
541:
505:
504:
486:
480:
479:
477:
476:
470:
463:
454:
445:
427:
293:
286:
282:
279:
273:
242:
234:
125:
118:
114:
111:
105:
103:
62:
38:
30:
21:
599:
598:
594:
593:
592:
590:
589:
588:
569:
568:
532:(10): 1013–30.
514:
509:
508:
501:
487:
483:
474:
472:
468:
461:
455:
448:
428:
424:
419:
401:
365:
330:
294:
283:
277:
274:
259:
243:
232:
208:
126:
115:
109:
106:
63:
61:
51:
39:
28:
23:
22:
15:
12:
11:
5:
597:
587:
586:
581:
567:
566:
560:
553:
552:
513:
512:External links
510:
507:
506:
499:
481:
446:
421:
420:
418:
415:
414:
413:
408:
400:
397:
396:
395:
376:
372:
364:
361:
329:
326:
296:
295:
246:
244:
237:
231:
228:
212:floating point
207:
204:
165:floating point
128:
127:
42:
40:
33:
26:
9:
6:
4:
3:
2:
596:
585:
582:
580:
577:
576:
574:
564:
561:
558:
555:
554:
549:
545:
540:
535:
531:
527:
526:
521:
516:
515:
502:
496:
492:
485:
467:
460:
457:Weiss, Alan.
453:
451:
443:
439:
437:
431:
426:
422:
412:
409:
406:
403:
402:
393:
389:
385:
381:
377:
373:
370:
369:
368:
360:
358:
353:
351:
347:
343:
339:
335:
325:
323:
319:
313:
309:
307:
303:
292:
289:
281:
271:
267:
263:
257:
256:
252:
247:This section
245:
241:
236:
235:
227:
224:
219:
217:
213:
203:
201:
197:
193:
189:
185:
181:
177:
173:
168:
167:performance.
166:
162:
158:
154:
150:
146:
142:
138:
134:
124:
121:
113:
102:
99:
95:
92:
88:
85:
81:
78:
74:
71: –
70:
66:
65:Find sources:
59:
55:
49:
48:
43:This article
41:
37:
32:
31:
19:
529:
523:
490:
484:
473:. Retrieved
441:
435:
425:
366:
363:Shortcomings
354:
341:
331:
314:
310:
299:
284:
275:
260:Please help
248:
222:
220:
209:
198:version for
188:indirections
169:
132:
131:
116:
107:
97:
90:
83:
76:
64:
52:Please help
47:verification
44:
357:clock rates
344:(Dhrystone
194:, with the
69:"Dhrystone"
573:Categories
475:2020-04-28
417:References
388:data cache
350:VAX 11/780
139:computing
80:newspapers
411:Geekbench
278:June 2021
249:does not
223:synthetic
216:Whetstone
161:Whetstone
141:benchmark
137:synthetic
133:Dhrystone
110:June 2021
466:Archived
444:rystone.
399:See also
322:CoreMark
180:ALGOL 68
548:9026014
438:etstone
392:SPECint
328:Results
270:removed
255:sources
178:, SAL,
172:FORTRAN
149:integer
94:scholar
546:
497:
407:(SPEC)
184:Pascal
182:, and
96:
89:
82:
75:
67:
544:S2CID
469:(PDF)
462:(PDF)
342:DMIPS
318:EEMBC
135:is a
101:JSTOR
87:books
18:DMIPS
495:ISBN
346:MIPS
338:CISC
336:vs.
334:RISC
253:any
251:cite
200:Unix
176:PL/1
73:news
534:doi
430:Pun
264:by
192:Ada
157:CPU
56:by
575::
542:.
530:27
528:.
522:.
464:.
449:^
432::
359:.
174:,
550:.
536::
503:.
478:.
442:h
436:h
434:w
291:)
285:(
280:)
276:(
272:.
258:.
196:C
155:(
123:)
117:(
112:)
108:(
98:·
91:·
84:·
77:·
50:.
20:)
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.
