| Number | 413
|
| Category | errata
|
| Synopsis | 4. scalar expressions
|
| State | lrmdraft
|
| Class | errata-discuss
|
| Arrival-Date | Aug 01 2003
|
| Originator | Shalom.Bresticker@motorola.com
|
| Release | 2001b: 4
|
| Environment |
|
| Description |
Section 4, para. 4, says "A scalar expression is an expression that evaluates to a scalar (single-bit) result. If the expression evaluates to a vector (multibit) result, then the least significant bit of the result is used as the scalar result." The paragraph has the following problems: - It contradicts itself. If it evaluates to a scalar, then it cannot evaluate to a vector. - So far as I can see, the phrase "scalar expression" is not used anywhere else in the LRM. - The only place I remember offhand where the described behavior applies is in @(posedge expression) and @(negedge expression). I think everywhere else, such as in "if (expr)" or expr1 in "expr1 ? expr2 : expr3", the behavior is to treat non-zero as true and zero as false. That behavior is certainly more common. I suggest to delete para. 4. -- Shalom Bresticker Shalom.Bresticker@motorola.com Design & Reuse Methodology Tel: +972 9 9522268 Motorola Semiconductor Israel, Ltd. Fax: +972 9 9522890 POB 2208, Herzlia 46120, ISRAEL Cell: +972 50 441478 |
| Fix |
Delete paragraph 4 in Clause 4 ("A scalar expression ..."). |
| Audit-Trail |
The original proposal was as follows: In paragraph 4 in Clause 4: "A scalar expression is an expression that evaluates to a scalar (single-bit) result. If the expression evaluates to a vector (multibit) result, then the least significant bit of the result is used as the scalar result." Delete sentence 2, leave only sentence 1. (Parenthetical note: I did find one more place which shows the behavior described in sectence 2: in the scalar_timing_check_condition in section 15.6, which is the subject of issue #237. There too, it says, "The conditioning signal shall be a scalar net; if a vector net or an expression resulting in a multi-bit value is used, then the least significant bit of the vector net or the expression value is used." I still propose to delete sentence 2, since it seems to be the exceptional behavior and not the usual behavior.) This proposal was modified by the ETF to delete the entire paragraph. |
| Unformatted |
|
Hosted by Boyd Technology