On Fri, 26 Dec 2025 at 06:43, Li-Hang Lin <[email protected]> wrote: > > Fix an error in rsqrte_f64() where the sign bit was being > placed incorrectly. Specifically, ensure f64_sign is deposited > into bit 63. > > Additionally, update the comments regarding the exponent (exp) bit > width to reflect the correct double-precision specifications.
This seems like it would produce incorrect results -- do you have an example of an instruction plus input data values that p produces a different output value to the hardware? thanks -- PMM
