		Search-engine friendly clone of the ACL2 documentation.

Oddexpr-check

Vl-expr-probable-selfsize

Heuristically estimate an expression's size.

Signature
(vl-expr-probable-selfsize x ss) → size
Arguments: x — Guard (vl-expr-p x).; ss — Guard (vl-scopestack-p ss).
Returns: size — The probable size of this expression, or nil for failure.
Type (maybe-natp size).

There's no reason to believe the size it returns will be the eventual size of the expression because size propagation hasn't been taken into account; in fact we may just fail and return nil for a number of reasons (for instance the wire ranges may not be resolved yet), there's no way to get the failure reason.

On the other hand, I think it should be the case that the final size of the expression will always be at least as much as this selfsize, if it returns a non-nil value. And we can use this before resolving ranges, etc., which makes it useful for simple linting.

Definitions and Theorems

Function: vl-expr-probable-selfsize

(defun vl-expr-probable-selfsize (x ss)
  (declare (xargs :guard (and (vl-expr-p x)
                              (vl-scopestack-p ss))))
  (let ((__function__ 'vl-expr-probable-selfsize))
    (declare (ignorable __function__))
    (b* (((mv & size)
          (vl-expr-selfsize x ss *fake-modelement* nil)))
      size)))

Theorem: maybe-natp-of-vl-expr-probable-selfsize

(defthm maybe-natp-of-vl-expr-probable-selfsize
  (b* ((size (vl-expr-probable-selfsize x ss)))
    (maybe-natp size))
  :rule-classes :type-prescription)