Re: Inliner behaviour - tiny changes lead to huge performance differences

On Fri, Nov 13, 2009 at 2:04 AM, Bryan O'Sullivan wrote:

And the lengthI function is defined more generally, in the hope that I could use it for both Int and Int64 lengths:

lengthI :: Integral a => Stream Char -> a lengthI (Stream next s0 _len) = loop_length 0 s0     where       loop_length !z s  = case next s of                            Done       -> z                            Skip    s' -> loop_length z s'                            Yield _ s' -> loop_length (z + 1) s' {-# INLINE[0] lengthI #-}

Would it help to SPECIALIZE lengthI for Int and Int64? -- Dave Menendez http://www.eyrie.org/~zednenem/