Changes between Version 3 and Version 4 of Commentary/CmmExceptions

Show
Ignore:
Timestamp:
01/05/07 13:16:35 (6 years ago)
Author:
p_tanski
Comment:

update on realToFrac (Infinity :: Float)

Legend:

Unmodified
Added
Removed
Modified

  • Commentary/CmmExceptions

    v3 v4  
    2121The following code will ''fail'' to produce a floating point exception or NaN on x86 machines (recall that 0.0/0.0 is NaN ''and'' a definite FPU exception): 
    2222 
    23 [in GHCi]: 
    24 Prelude> realToFrac (0/0 :: Float) :: Double 
     23[in GHCi-6.6 on PowerPC, OS X]: 
     24{{{ 
     25Prelude> 0.0/0.0 
     26NaN 
     27 
     28Prelude> realToFrac (0.0/0.0) :: Double 
     29Infinity 
     30 
     31Prelude> realToFrac (0.0/0.0 :: Float) 
     325.104235503814077e38 
     33 
     34Prelude> realToFrac (0.0/0.0 :: Float) :: Double 
     355.104235503814077e38 
     36 
     37Prelude> realToFrac (1.0/0.0) 
     38Infinity 
     39Prelude> realToFrac (1.0/0.0 :: Float) 
     403.402823669209385e38 
     41 
     42}}} 
     43 
     44This bug is not due to the lack of FPU exceptions in Cmm but bears mention as the internal conversion performed in 'realToFrac' on 'Float's and might benefit from FPU exceptions.