[logo] 
 
Home
News
Activity
About/Contact
Major Tools
  Dinotrace
  Verilator
  Verilog-mode
  Verilog-Perl
Other Tools
  BugVise
  CovVise
  Force-Gate-Sim
  Gspice
  IPC::Locker
  Rsvn
  SVN::S4
  SystemPerl
  Voneline
  WFH
General Info
  Papers

Issue #478

%Error: ...: Internal: Blocking <= assignment in non-clocked block, should have converted in V3Active

Added by Alex Solomatnikov about 1 year ago. Updated 10 months ago.

Status:Closed Start date:04/05/2012
Priority:Normal Due date:
Assignee:Wilson Snyder % Done:

0%

Category:Unsupported
Target version:-

Description

Source code looks like:

    initial
    begin
...
        a1 <= 1'b0;
        a2 <= 0;
        a3 <= 0;
        a4 <= 0;
        a5 <= 1'b0;
        a6 <= 1'b1;
        a7 <= 1'b1; 
        a8 <= 1'b1;                 
        a9 <= 1'b0;
        a10 <= 1'b0;
        a11 <= 0;

verilator complains about these assignments.

History

Updated by Alex Solomatnikov about 1 year ago

There is also an error message in the same initial block about non-blocking assignment to an array

%Error-BLKLOOPINIT: ...: Unsupported: Delayed assignment to array inside for loops (non-delayed is ok - see docs)

Updated by Alex Solomatnikov about 1 year ago

After I tried to convert non-blocking assignments <= to blocking =, I get:

%Error-BLKANDNBLK: ...: Unsupported: Blocked and non-blocking assignments to same variable: ...

Apparently, in this behavioral model some state registers are initialized using initial block. There is no way to compile this model with verilator because non-blocking assignment in initial block causes internal error and blocking assignment in initial block causes this error.

Updated by Wilson Snyder about 1 year ago

In the BLKANDNBLK case, even though these are errors you can turn them off like you would a warning; this is because Verilator won't be modeling them as would be a normal simulator that honors the <= delays. Probably both <= and = should be allowed in initial blocks without warnings.

Updated by Alex Solomatnikov about 1 year ago

My last message was wrong and should be disregarded: in fact Error-BLKANDNBLK is related to different module.

Updated by Wilson Snyder about 1 year ago

  • Category set to Unsupported
  • Status changed from New to Resolved
  • Assignee set to Wilson Snyder

Added a new INITIALDLY warning you can suppress if you wish.

Fixed in git towards 3.840++.

Updated by Wilson Snyder 10 months ago

  • Status changed from Resolved to Closed

In 3.840.

Also available in: Atom