[libre-riscv-dev] div/mod algorithm written in python

Luke Kenneth Casson Leighton lkcl at lkcl.net
Tue Jul 9 15:05:46 BST 2019


jacob: this is what i get on running test_core.py.

lkcl at fizzy:~/src/libreriscv/ieee754fpu/src$ nosetests3
ieee754/div_rem_sqrt_rsqrt/test_core.py
EEE
======================================================================
ERROR: test_bit_width_8_fract_width_4_radix_2
(ieee754.div_rem_sqrt_rsqrt.test_core.TestDivPipeCore)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/lkcl/src/libreriscv/ieee754fpu/src/ieee754/div_rem_sqrt_rsqrt/test_core.py",
line 228, in test_bit_width_8_fract_width_4_radix_2
    log2_radix=1))
  File "/home/lkcl/src/libreriscv/ieee754fpu/src/ieee754/div_rem_sqrt_rsqrt/test_core.py",
line 194, in handle_case
    traces=[*dut.traces()]) as sim:
  File "/home/lkcl/src/riscv/nmigen/nmigen/back/pysim.py", line 517,
in __enter__
    var_init = signal.decoder(signal.reset).replace(" ", "_")
AttributeError: 'DivPipeCoreOperation' object has no attribute 'replace'


On Tue, Jul 9, 2019 at 12:29 PM Luke Kenneth Casson Leighton
<lkcl at lkcl.net> wrote:
>
>
>
> On Tuesday, July 9, 2019, Jacob Lifshay <programmerjake at gmail.com> wrote:
>>
>> found one of the bugs causing simulation to fail:
>> https://github.com/m-labs/nmigen/issues/141
>
>
> Ah excellent.
>
> Also i just remembered where I know bigfloat from, The de Vries formula for the fine structure constant. http://lkcl.net/reports/fine_structure_constant/
>
> I used bigfloat there, it's excellent. Good choice, jacob.
>
> Now how do we fit 32 and 64 bit random ints into it (direct conversion)
>
> L.
>
>
>
> --
> ---
> crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
>



More information about the libre-riscv-dev mailing list