Glibc log2 Function Up To 69% Faster Thanks To FMA
Intel engineer and open-source compiler expert H.J. Lu has landed a much faster log2() implementation within the GNU C Library (glibc) that makes use of FMA instructions with modern x86_64 CPUs.
H.J. Lu has committed an FMA-optimized log2() function to the GNU C Library for obtaining the base 2 logarithm of a value.
Using an older Intel Skylake CPU for reference, his Fused Multiply Add (FMA) optimized log2 function was 34% faster or up to 69% faster for maximum performance.
It's yet another nice hand-tuned function for the GNU C Library that's now in Glibc Git.
H.J. Lu has committed an FMA-optimized log2() function to the GNU C Library for obtaining the base 2 logarithm of a value.
Using an older Intel Skylake CPU for reference, his Fused Multiply Add (FMA) optimized log2 function was 34% faster or up to 69% faster for maximum performance.
It's yet another nice hand-tuned function for the GNU C Library that's now in Glibc Git.
13 Comments