asr doesn't work at all

[Kuree]

When I was running Jeff's end-to-end tests, based on the waveform I've seen asr is not implemented correctly in RTL. See the waveform below.

Then I went ahead and poked around inside lassen. Even if I added asr to RTL, the tests is still passing. That's odd. I then ask pytest to print out everything, and here are the things I found:

1. asr is not working
2. fault is not reporting error even though verilator fails.
3. (maybe a problem) where is the SMT formal check on the RTL? Or this is not how formal works? (please forgive me if I make a wrong assertion.)

Here is the link to the print out: https://travis-ci.org/StanfordAHA/lassen/builds/532629001#L1217
You can reproduce the individual test using test_asr branch with the following command:

pytest tests/test_rtl.py -k "test_rtl[False-mode0-asr]" -s

You should see the verilator assertion failed but pytest passed.

I'd call for a thorough manually examination on every single op we have a manually test them to prevent these kinds of mistake from happening again.

[leonardt]

Fault missing the error is from a regression introduced recently (good news because that means that errors could only have slipped through due to recent changes, which I hope there haven't been many).

This PR (leonardt/fault#88) changed the logic to redirect stdout to a file (https://github.com/leonardt/fault/blob/7e117e56c4e414631bb1c40c1e1f28848f9cedc3/fault/verilator_target.py#L379) but a side-effect of that change was that the exit code for the first command in the pipe is now dropped (which means fault misses the assertion).

Working on a fix now (will also add a test that makes sure that an assertion occurs for a failing test so we catch issues like this in the future).

Not sure what the issue with asr is yet though.

[Kuree]

asr is not padding signed bits in RTL. The functional model is correct though. I will take a look at the RTL today to see what's wrong there.

EDIT: Can you just throw an exception in verilator? That should definitely stop the testing and report an error.

[leonardt]

This fixes the verilator silent failure issue: leonardt/fault#100 by setting -o pipefail before the verilator run command, investigating the RTL bug now, I suspect it's because it's generating a logical shift from mantle (instead of an arithmetic shift).

[leonardt]

First issue was that mantle was not setting the __rshift__ operator on the SIntType to be asr (so it defaulted to LSR inherited from the BitsType, fixed by phanrahan/mantle#147

Test is still failing, I suspect due to a polymorphism (lack of magma support) issue, investigating this now

[leonardt]

Tests is passing locally for me now, opened a PR #69, but depends on some upstream branches being merged.

[Kuree]

Thanks for the quick fix!