Unexpected inversion when using inverted registered outputs #12
Comments
and3rson
changed the title
|
FYI - GALasm maintainers confirmed that this is indeed a bug (see link in the previous post.) |
|
Hi. Sorry for not getting back to you earlier - I was off github for a while. I'm slightly nervous about changing the behaviour if anyone's relying on it, even if it's wrong, but given that galasm has applied a fix, I reckon I can change this behaviour without worrying about breaking bug compatibility. I will fix this when I have a moment. Thanks! |
simon-frankau
added a commit
that referenced
this issue
Mar 7, 2024
As pointed out in #12 , the GAL22V10 in registered mode is unusual - the feedback value is always inverted, even if the output pin value is not. This was not captured correctly in the code. This is hopefully fixed, along with some new tests to keep an eye on this annoying corner case.
|
I believe this should now be fixed in b1be5ae . If you have a moment to take a look, I'd appreciate it. If you see any issues, please do let me know. Cheers. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I've originally posted this issue in GALasm repo (daveho/GALasm#16 (comment)), but I can also reproduce this behavior in galette.
TL;DR: Consider the following equations:
When assembled with
galasm(orgalette) and then disassembled withjed2eqn, I get the following equations:This seems completely wrong, since Q2 clearly should be equal to Q1. Any thoughts?
The text was updated successfully, but these errors were encountered: