Skip to content

Conversation

@kulp
Copy link
Owner

@kulp kulp commented Jul 12, 2021

In the process of introducing Verilator support (#48), I found some minor issues that Verilator rightly complains about. They are collected here.

kulp added 5 commits July 3, 2021 16:04
Although it is true that CI support was added for non-x86 architectures,
inluding in #36, the Travis-CI implementation of that CI was subsequently
removed (#72), so it does not make sense to notate the addition in the
Changelog for v0.9.8.
The amount of entropy tested by the `@` op-test is pretty low, and I have
observed on multiple occasions the test passing even when the hardware is not
doing its job. The current commit strengthens the randomization test a bit. An
even stronger test might ensure that a sole one or zero in a word of zeros or
ones is being extracted successfully.
The purpose of the Bit function, as it was introduced in commit
eda7dc4, was to keep columnar alignment for
readability. As of commit f98aba7, the
contained expression shrank significantly, but could not be inlined directly
since its result width was too wide. Now I recognize that I should have been
using bit-indexing instead of bit-shifting anyway, and the silly function can
be removed entirely.

As a happy side-effect, this change addresses a bit-width lint noted by
Verilator.
@codecov

This comment has been minimized.

@kulp kulp marked this pull request as ready for review July 12, 2021 22:53
@kulp kulp merged commit c0a4ee4 into develop Jul 12, 2021
@kulp kulp deleted the hw-fixes branch July 12, 2021 22:54
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants