Berkeley's riscv-torture is dead?

[jhause...@gmail.com]

It appears to me that U.C. Berkeley's RISC-V Torture Test Generator,
found here:
https://github.com/ucb-bar/riscv-torture
is not being maintained, and has now bit-rotted to the point of
unusability, thanks to constant churn in SBT (the Scala Build Tool)
on which it depends.  At least, that's been my experience with simply
cloning the repository and following the instructions.  I first ran
into Issue #22:
https://github.com/ucb-bar/riscv-torture/issues/22
which has been sitting unresolved for almost a year.  My inexpert
attempt to patch for that bug only got me as far as Issue #25:
https://github.com/ucb-bar/riscv-torture/issues/25
which also hasn't gotten any attention since being filed nine months
ago.

Is anybody actually using 'riscv-torture' any more?  If so, how?
If not, what is the publicly available replacement that I don't know
about?

Thanks,

    - John Hauser

[Jerry Zhao]

Torture is maintained when someone finds it useful, which is less and less frequent these days.
I opened a PR to bump SBT in torture, it should compile again with these changes.
https://github.com/ucb-bar/riscv-torture/pull/29

An alternative is Google's riscv-dv instruction generator. https://github.com/google/riscv-dv .
When I tried this years ago, it was not significantly more capable than torture, but they seem to 
have added many new features recently.

-Jerry Zhao

--
You received this message because you are subscribed to the Google Groups "RISC-V HW Dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hw-dev+un...@groups.riscv.org.
To view this discussion on the web visit https://groups.google.com/a/groups.riscv.org/d/msgid/hw-dev/39dc0ce4-1f26-458b-9274-ff1960f0a623n%40groups.riscv.org.

[jhause...@gmail.com]

Jerry Zhao wrote:
> Torture is maintained when someone finds it useful, which is less and
> less frequent these days.  I opened a PR to bump SBT in torture, it
> should compile again with these changes.
> https://github.com/ucb-bar/riscv-torture/pull/29

Thank you, Jerry, that solved it for me!  I would not have been able to
fix the problem so quickly myself.

Thanks also for the tip about Google's instruction generator.  I'll
give it a try sometime.

    - John Hauser

[Neel Gala]

Shakti's aapg tool is equally powerful. Recently features to generate self-checking tests (via checksum) have also been added. Its actively maintained as well.

Do check out the wiki for more information on the features.

--
You received this message because you are subscribed to the Google Groups "RISC-V HW Dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hw-dev+un...@groups.riscv.org.

To view this discussion on the web visit https://groups.google.com/a/groups.riscv.org/d/msgid/hw-dev/fada22e7-8d83-44d7-a08b-217e101d2274n%40groups.riscv.org.

--

Neel Gala

[Josh Scheid]

On Tue, May 4, 2021 at 9:12 PM Neel Gala <neel...@gmail.com> wrote:

Shakti's aapg tool is equally powerful. Recently features to generate self-checking tests (via checksum) have also been added. Its actively maintained as well.

Do check out the wiki for more information on the features.

I'll second the recommendation to try aapg for a libre+gratis generator.  It was very easy to get started with and I'm happy to see ongoing maintenance.  google/risc-dv still requires a commercial simulator at this point.

-Josh