Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Fix a double-rounding bug in fma{,f,l}. The bug would occur in | das | 2011-10-15 | 1 | -9/+29 |
| | | | | | | | | | | round-to-nearest mode when the result, rounded to twice machine precision, was exactly halfway between two machine-precision values. The essence of the fix is to simulate a "sticky bit" in the pathological cases, which is how hardware implementations break the ties. MFC after: 1 month | ||||
* | Include math.h for the fmaf() prototype. | das | 2008-03-29 | 1 | -0/+2 |
| | |||||
* | Add fma() and fmaf(), which implement a fused multiply-add operation. | das | 2005-01-22 | 1 | -0/+47 |