Issue #320
Interface port connection using .* is wrong (?)
| Status: | Assigned | Start date: | 01/27/2011 | |
|---|---|---|---|---|
| Priority: | Normal | Due date: | ||
| Assignee: | Wilson Snyder | % Done: | 0% |
|
| Category: | Autos | |||
| Target version: | - |
Description
Hi,
It seems to me there are at least 2 problems with the .* for interfaces. E.g.
Second, I'm not sure the internal declarations (i.e. not the port declarations) should be in the interface connection. I think it should not since these are not interface ports. When I try to compile, it produces compilation error. E.g. Signals 'a' and 'b' for c interface instantiation in 'top' should not be in the interface connection, same for signal 'e' for 'f' interface instantiation in 'top'.
Thanks,
Pierre
It seems to me there are at least 2 problems with the .* for interfaces. E.g.
interface intf1; logic a; logic b; modport mp1 (input a, output b); endinterface interface intf2 (intf1.mp1 c); logic e; endinterface module top; intf1 c(.*); intf2 f(.*); endmoduleFirst, some comments are left out when saving the file after C-c C-a (this is cosmetics, but is a little annoying).
Second, I'm not sure the internal declarations (i.e. not the port declarations) should be in the interface connection. I think it should not since these are not interface ports. When I try to compile, it produces compilation error. E.g. Signals 'a' and 'b' for c interface instantiation in 'top' should not be in the interface connection, same for signal 'e' for 'f' interface instantiation in 'top'.
Thanks,
Pierre
History
Updated by Wilson Snyder over 2 years ago
- Status changed from New to Assigned
#2 needs more thought, right now it's a side effect of how AUTOINST works, which simply isn't right.
I just fixed 1, I can make a release now if you want it.
Also available in: Atom
![[logo]](/img/veripool_small.png)