Possible std.math.exp() bug

[Don Clugston]

Walter Bright wrote:
> Don Clugston wrote:
>> However, there *is* an accuracy issue in tgamma, which occurred when 
>> it was translated from D back to C (!) for the DMC compiler.
> 
> I thought that was fixed? Can you redownload snn.lib?

When it was first included, it had a major bug in lgamma and a minor 
error in tgamma. The bug in lgamma() was fixed in the subsequent 
release. Last time I checked, tgamma was faulty (although it's subtle -- 
about the fifth digit is wrong).

I just tried to run my test program, but mathextra now gives an internal 
error (ztc.type.c 308) -- new regression in DMD 0.151, I'll track it down.