From 7d6b391572e5ce694558ce663853664d03b70805 Mon Sep 17 00:00:00 2001
From: Stephen Williams <steve@icarus.com>
Date: Sat, 13 Sep 2008 17:40:19 -0700
Subject: [PATCH] In continuous assignment, r-value is sized or lossless, never
 self determined.

The r-value expression takes on at least the width of the l-value. If
there is a sized width greater then the l-value width, take that on
instead. But if the r-value is unsized, there are unsized sub-expressions
that cause the expression as a whole to be unlimited width, so elaborate
and evalulate the r-value as lossless.
---
 elaborate.cc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/elaborate.cc b/elaborate.cc
index a2cb6e6c0..4e5b40220 100644
--- a/elaborate.cc
+++ b/elaborate.cc
@@ -110,7 +110,7 @@ void PGAssign::elaborate(Design*des, NetScope*scope) const
 		 << ", unsized_flag=" << (unsized_flag?"true":"false") << endl;
       }
 
-      int expr_wid = unsized_flag? -1 : (int) use_width;
+      int expr_wid = unsized_flag? -2 : (int) use_width;
       NetExpr*rval_expr = elab_and_eval(des, scope, pin(1),
 					expr_wid, lval->vector_width());