News for hackers. Stuff that matters.
Re:Haskell Beats C Using Generalized Stream Fusion [pdf] by 33a on Sunday April 07, 2013 @04:46PM (#5508678)
If Haskell beats C on that loop, then that is a compiler bug. Did they try repeating these experiments using say icc?
Re:Haskell Beats C Using Generalized Stream Fusion [pdf] by kevingadd on Sunday April 07, 2013 @05:28PM (#5508851)
Personally, I would say that leaping straight to 'compiler bug' is a little unwarranted. If you really think it's a compiler bug, you should compile the code (they provided it) using the settings they used (I think they provided that too; you can see them in a comment in this thread, actually) and then look at the output assembly and find the bug and file a bug against the compiler.
Re:Haskell Beats C Using Generalized Stream Fusion [pdf] by lallysingh on Sunday April 07, 2013 @06:57PM (#5509096)
Or Haskell semantics allows more aggressive optimization than C.

icc's not actually that much better than gcc in optimization, it's primarily a much better runtime library that gives icc its speed boost.