[Libre-soc-dev] Simulator speed status

[Lauri Kasanen]

On Tue, 18 May 2021 11:23:22 +0100
Luke Kenneth Casson Leighton <lkcl at lkcl.net> wrote:

> after investigation, i had in
> fact updated local nmigen (12th april 2021) and i've just pushed that with
> a forced master update

I'm now able to run the fp tests successfully.

53s for 10 tests / 5.3s per 1-2 instructions is very slow, I
probably won't even try the other existing tests at this speed. And that
was with >/dev/null to cut the massive prints out.

Hm.. If that's ~3.5s per instruction, for a 700 instruction test case,
without loops, execution would take ~41 minutes (!). Is that expected
speed?

If that's all expected, I can't work with an hour per iteration. A
compiled simulator would be a blocker then. I realize this isn't
something Luke can do, not being on his critical path. Is there funding?

I'm wary of volunteering, since the existing simulator is python which
I don't do. Converting it to C/C++ might take me quite a long time,
making it unviable. Getting new volunteers doesn't appear easy either.

- Lauri