* When used with -tempinduct mode, -seq <N> causes assertions to be
ignored in the first N steps. While this has uses for reset modelling,
for these test cases it is unnecessary and could lead to failures
slipping through uncaught
Commit c8e0ac0 introduces a regression on handling case exprs that look
like sva labels. After some debugging, we shouldn't push the identifier
ast node to the ast_stack, otherwise, we will get the following
assertion failure:
```
➜ /tmp yosys -p 'read -sv a1.v'
/----------------------------------------------------------------------------\
| yosys -- Yosys Open SYnthesis Suite |
| Copyright (C) 2012 - 2025 Claire Xenia Wolf <claire@yosyshq.com> |
| Distributed under an ISC-like license, type "license" to see terms |
\----------------------------------------------------------------------------/
Yosys 0.57+1 (git sha1 baa61a146, clang++ 20.1.8 -fPIC -O3)
-- Running command `read -sv a1.v' --
1. Executing Verilog-2005 frontend: a1.v
Parsing SystemVerilog input from `a1.v' to AST representation.
ERROR: Assert `extra->ast_stack.size() == 1' failed in frontends/verilog/verilog_parser.y:709.
➜ /tmp cat a1.v
module test(input wire A);
localparam TEST = 1;
always_comb begin
case (A)
TEST: assert(1);
endcase
end
endmodule
```
We encountered this issue before but with a different error message[^1],
[^1]: https://github.com/YosysHQ/yosys/issues/862
When FILTERLIB is defined (attempts to compile libparse more or less standalone,) mark the `LibertyParser::error()` as weak so utilities using libparse as a library can override its behavior (the default behavior being exit(1)). As the code is quite performance-critical, I've elected to not modify it to raise an exception or have a callback or similar and simply allow for a link-time replacement.
We're currently on version 3.6 of bison at Google, and Yosys
still correctly builds with it. This should better reflect
the actual requirements rather than an overly restrictive
check. If features from 3.8 are required it seems like bumping
would be appropriate.
Signed-off-by: Ethan Mahintorabi <ethanmoon@google.com>
