x86 Intrinsics for fma in Math Library

Andrew Haley aph at redhat.com
Thu Jul 21 16:26:35 UTC 2016

On 20/07/16 03:04, Joseph D. Darcy wrote:
> The existing fma tests are far from exhaustive and would benefit from 
> being augmented by additional test cases, in particular some test cases 
> which exercise the harder rounding cases for a hardware implementation.

I have been using one which excludes double rounding, and I can add it
to the JDK test.  Based on the logic here:


The big problem with the JDK test is that it doesn't exercise the
compilers.  In many ways it's a good test otherwise.


More information about the hotspot-compiler-dev mailing list