在 2022/6/7 17:11, g...@m-labs.hk 写道:
Attached is a patch to fix that.
The approach looks credible. However:
The proposed solution transforms calculation of `pow(x, y)`, where `y` is negative, into `1 / pow(x,
-y)`. This is not strictly equivalent. The maximum value that a `double` may repre
Dear Sir/Madam,
As mentioned in here, there is a floating point imprecision problem in
the pow function.
Originally,
#include
#include
int main() {
union {
double d;
unsigned long long int i;
} cast;
cast.d = pow(10., -9);
printf("%llx\n", cast.i);
return 0;
}
3e112e0be826d699
the probl
