Remove eval_expr during PEBinary elaboration
Turns out that call to eval_expr is redundant, and in degenerate situations can lead to so many wasteful calls to eval_expr that elaboration performance suffers dramatically.
This commit is contained in:
parent
6684b2db04
commit
d8f4ff4b57
|
|
@ -285,11 +285,6 @@ NetExpr* PEBinary::elaborate_expr(Design*des, NetScope*scope,
|
||||||
return 0;
|
return 0;
|
||||||
}
|
}
|
||||||
|
|
||||||
// If either expression can be evaluated ahead of time, then
|
|
||||||
// do so. This can prove helpful later.
|
|
||||||
eval_expr(lp, l_width);
|
|
||||||
eval_expr(rp, r_width);
|
|
||||||
|
|
||||||
return elaborate_expr_base_(des, lp, rp, expr_wid);
|
return elaborate_expr_base_(des, lp, rp, expr_wid);
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
|
||||||
Loading…
Reference in New Issue