#149: missed CSE opportunity
-----------------------------------------+----------------------------------
  Reporter:  nobody                      |          Owner:                  
      Type:  bug                         |         Status:  new             
  Priority:  normal                      |      Milestone:  _|_             
 Component:  Compiler                    |        Version:  5.04.2          
Resolution:  None                        |       Keywords:  optimisations   
  Testcase:  simplrun006                 |      Blockedby:                  
Difficulty:  Moderate (less than a day)  |             Os:  Unknown/Multiple
  Blocking:                              |   Architecture:  Unknown/Multiple
   Failure:  Runtime performance bug     |  
-----------------------------------------+----------------------------------
Changes (by michalt):

 * cc: michal.terepeta@… (added)


Comment:

 I've run the fixed benchhmark on GHC 7.0.2 and the differences are quite
 small:
 {{{
 warming up
 estimating clock resolution...
 mean is 5.608272 us (160001 iterations)
 found 2835 outliers among 159999 samples (1.8%)
   2620 (1.6%) high severe
 estimating cost of a clock call...
 mean is 41.08527 ns (43 iterations)
 found 4 outliers among 43 samples (9.3%)
   2 (4.7%) high mild
   2 (4.7%) high severe

 benchmarking playerMostOccur
 collecting 100 samples, 1 iterations each, in estimated 635.0040 ms
 bootstrapping with 100000 resamples
 mean: 3.506386 ms, lb 3.463573 ms, ub 3.627819 ms, ci 0.950
 std dev: 335.3037 us, lb 107.8520 us, ub 701.8364 us, ci 0.950
 found 3 outliers among 100 samples (3.0%)
   2 (2.0%) high severe
 variance introduced by outliers: 1.000%
 variance is unaffected by outliers

 benchmarking playerMostOccur'
 collecting 100 samples, 2 iterations each, in estimated 720.7155 ms
 bootstrapping with 100000 resamples
 mean: 3.606333 ms, lb 3.600894 ms, ub 3.616024 ms, ci 0.950
 std dev: 36.14909 us, lb 22.68614 us, ub 52.87152 us, ci 0.950
 found 10 outliers among 100 samples (10.0%)
   9 (9.0%) high severe
 variance introduced by outliers: 0.990%
 variance is unaffected by outliers

 benchmarking playerMostOccur
 collecting 100 samples, 67 iterations each, in estimated 561.6000 ms
 bootstrapping with 100000 resamples
 mean: 84.15780 us, lb 84.03386 us, ub 84.47981 us, ci 0.950
 std dev: 953.0503 ns, lb 438.4662 ns, ub 1.977162 us, ci 0.950
 found 7 outliers among 100 samples (7.0%)
   3 (3.0%) high mild
   4 (4.0%) high severe
 variance introduced by outliers: 0.990%
 variance is unaffected by outliers

 benchmarking playerMostOccur'
 collecting 100 samples, 70 iterations each, in estimated 566.6148 ms
 bootstrapping with 100000 resamples
 mean: 81.52763 us, lb 81.32001 us, ub 81.96656 us, ci 0.950
 std dev: 1.476358 us, lb 803.8508 ns, ub 2.630875 us, ci 0.950
 found 11 outliers among 100 samples (11.0%)
   4 (4.0%) high mild
   7 (7.0%) high severe
 variance introduced by outliers: 0.995%
 variance is unaffected by outliers

 benchmarking playerMostOccur
 collecting 100 samples, 1 iterations each, in estimated 923.5859 ms
 bootstrapping with 100000 resamples
 mean: 7.817296 ms, lb 7.790770 ms, ub 7.877817 ms, ci 0.950
 std dev: 193.8747 us, lb 70.80833 us, ub 338.0659 us, ci 0.950
 found 4 outliers among 100 samples (4.0%)
   3 (3.0%) high severe
 variance introduced by outliers: 0.997%
 variance is unaffected by outliers

 benchmarking playerMostOccur'
 collecting 100 samples, 1 iterations each, in estimated 757.6227 ms
 bootstrapping with 100000 resamples
 mean: 7.514011 ms, lb 7.495620 ms, ub 7.552087 ms, ci 0.950
 std dev: 130.1032 us, lb 72.61251 us, ub 236.9545 us, ci 0.950
 found 6 outliers among 100 samples (6.0%)
   4 (4.0%) high mild
   2 (2.0%) high severe
 variance introduced by outliers: 0.995%
 variance is unaffected by outliers

 benchmarking playerMostOccur
 collecting 100 samples, 1 iterations each, in estimated 82.58250 s
 bootstrapping with 100000 resamples
 mean: 777.5068 ms, lb 776.6123 ms, ub 778.8386 ms, ci 0.950
 std dev: 5.512690 ms, lb 4.044142 ms, ub 7.447603 ms, ci 0.950
 found 10 outliers among 100 samples (10.0%)
   4 (4.0%) high mild
   6 (6.0%) high severe
 variance introduced by outliers: 0.990%
 variance is unaffected by outliers

 benchmarking playerMostOccur'
 collecting 100 samples, 1 iterations each, in estimated 76.42000 s
 bootstrapping with 100000 resamples
 mean: 748.9900 ms, lb 747.7942 ms, ub 750.6694 ms, ci 0.950
 std dev: 7.211566 ms, lb 5.591098 ms, ub 9.302462 ms, ci 0.950
 found 9 outliers among 100 samples (9.0%)
   3 (3.0%) high mild
   6 (6.0%) high severe
 variance introduced by outliers: 0.990%
 variance is unaffected by outliers
 }}}
 It's interesting that the times are not the same. On the other hand, it
 seems
 that the original issue has been fixed.

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/149#comment:19>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler

_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs

Reply via email to