[libre-riscv-dev] [Bug 354] New: idea/optimisation: make FP CMP also do INT CMP

[libre-riscv-dev] [Bug 354] New: idea/optimisation: make FP CMP also do INT CMP

[bugzilla-daemon]

https://bugs.libre-soc.org/show_bug.cgi?id=354

            Bug ID: 354
           Summary: idea/optimisation: make FP CMP also do INT CMP
           Product: Libre-SOC's first SoC
           Version: unspecified
          Hardware: PC
                OS: Linux
            Status: CONFIRMED
          Severity: enhancement
          Priority: ---
         Component: Source Code
          Assignee: lkcl@lkcl.net
          Reporter: lkcl@lkcl.net
                CC: libre-riscv-dev@lists.libre-riscv.org
   NLnet milestone: ---

from mitch alsup on comp.arch:

This all falls out for FREE if you start with FP compare to ZERO and
then figure out how the same logic can be used for integer instead of
the other way around.

I have the logic schematic (3 pages) is you desire to peruse. The overhead
of adding FP to INT compare is around 10% in gate count.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

[libre-riscv-dev] [Bug 354] idea/optimisation: make FP CMP also do INT CMP

[bugzilla-daemon]

https://bugs.libre-soc.org/show_bug.cgi?id=354

Luke Kenneth Casson Leighton <lkcl@lkcl.net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|Source Code                 |ALU (including IEEE754
                   |                            |16/32/64-bit FPU)

-- 
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev