Changes between Version 24 and Version 25 of DataParallel/BenchmarkStatus


Ignore:
Timestamp:
Mar 5, 2009 12:04:23 PM (5 years ago)
Author:
chak
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • DataParallel/BenchmarkStatus

    v24 v25  
    2121 
    2222|| '''Program''' || '''Problem size''' || '''sequential''' || '''P=1''' || '''P=2''' || '''P=4''' || '''P=8''' || 
     23|| !SumSq, primitives || 10M || 22 || 40 || 20 || 10 || 5 || 
     24|| !SumSq, vectorised || 10M || 22 || 292 || 170 || 119 || 171 || 
     25|| !SumSq, ref C ||10M || 9 || – || – || – || – || 
    2326|| DotP, primitives || 100M elements || 823/823/824 || 812/813/815 || 408/408/409 || 220/223/227 || 210/214/221 || 
    2427|| DotP, vectorised || 100M elements || 823/824/824 || 814/816/818 || 412/417/421 || 222/225/227 || 227/232/238 || 
     
    3639 * We need an extra -funfolding-use-threshold.  We don't really want users having to worry about that. 
    3740 * `mapP (\x -> x * x) xs` essentially turns into `zipWithU (*) xs xs`, which doesn't fuse with `enumFromTo` anymore.  We have a rewrite rule in the library to fix that, but that's not general enough.  We really would rather not vectorise the lambda abstraction at all. 
     41 * `enumFromTo` doesn't fuse due to excessive dictionaries in the unfolding of `zipWithUP`. 
    3842 
    3943==== Comments regarding DotP ====