Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

AUTOLOGIC/AUTOINST for unpacked array signal input to submodule #1453

Open
veripoolbot opened this issue Jun 2, 2019 · 3 comments
Open

AUTOLOGIC/AUTOINST for unpacked array signal input to submodule #1453

veripoolbot opened this issue Jun 2, 2019 · 3 comments

Comments

@veripoolbot
Copy link
Collaborator


Author Name: David Rogoff
Original Redmine Issue: 1453 from https://www.veripool.org


Hi Wilson.

Thanks for the super-quick answer to my last question!

Here's another:

I have a signal like this:
logic [7:0] sig_a [2];

sig_a [0] 

comes from one instance of module xx and sig_a[1] from the other instance.

The full ```sig_a [2]


When I expand autos, I get this under /*AUTOLOGIC*/

  logic [0] [7:0] sig_a; // From inst_xx_0

If it's the other direction, it works fine, but it looks like AUTO prioritizes outputs of modules and gets confused.  How can I fix this without manually declaring all of these signals?

Thanks,

David


@veripoolbot
Copy link
Collaborator Author


Original Redmine Comment
Author Name: Wilson Snyder (@wsnyder)
Original Date: 2019-06-04T00:53:39Z


Verilog-mode doesn't figure out how to deal with 2-D and up structures, it just assumes the first output is what you want. The combining code is already complicated so is unlikely to get reworked to improve this case, so you'll need to declare it manually.

@veripoolbot
Copy link
Collaborator Author


Original Redmine Comment
Author Name: David Rogoff
Original Date: 2019-10-11T17:14:22Z


Wilson, I understand you not wanting to fix this but I keep hitting this issue on huge upper-level integration modules - exactly where AUTOs are great. Have you considered making a paid version of verilog-mode to give you and other developers incentive to do some of the more difficult fixes? After years and years of relying on your code (and emacs) I'm starting to look at eclipse with plugins like DVT (which is certainly not free!).

David

@veripoolbot
Copy link
Collaborator Author


Original Redmine Comment
Author Name: Wilson Snyder (@wsnyder)
Original Date: 2019-10-18T02:24:29Z


If you can provide some examples I'll take a look if there's some straightforward cases I can cover (gratis). Please include a reasonable number of cases in a file along with how you think they should be properly combined, including some cases that shouldn't combine. Note presently all the following are different internally so each of these need to be tested against whatever combine permutations:

reg [x:y] packed;
reg [x:y][x:y] packed2d;
reg unpacked [x:y];
reg unpacked2d [x:y][x:y];
reg [x:y] packedunpacked [x:y];
reg [x:y][x:y] unpacked2d [x:y][x:y];

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant