Trevor Gross ba0cfe58dd Run icount benchmarks once with softfloat and once with hardfloat ... These benchmarks are fast to run, so the time cost here is pretty minimal. Running softfloat benchmarks just ensures that we don't e.g. test the performance of `_mm_sqrt_ss` rather than our implementation, and running without softfloat gives us a way to see the effect of arch intrinsics.		2025-01-21 07:58:05 +00:00
..
compiler-builtins/libm	Run icount benchmarks once with softfloat and once with hardfloat	2025-01-21 07:58:05 +00:00