math: fix i386/expl.s with more precise x*log2e

[musl] / src / math / tgammaf.c

#include <math.h>

float tgammaf(float x)
{
        return tgamma(x);
}