| Number | 526
|
| Category | errata
|
| Synopsis | 7.8: "conflicting strength is ignored" ??
|
| State | lrmdraft
|
| Class | errata-discuss
|
| Arrival-Date | Jan 06 2004
|
| Originator | Shalom.Bresticker@motorola.com
|
| Release | 2001b: 7.1.2, 7.8
|
| Description |
7.8 "pullup and pulldown sources" says, "A pullup source shall place a logic value 1 on the nets connected in its terminal list. A pulldown source shall place a logic value 0 on the nets connected in its terminal list. The signals that these sources place on nets shall have pull strength in the absence of a strength specification. If conflicting strength specification is declared, it shall be ignored. There shall be no delay specifications for these sources." What does "conflicting strength specification" mean ? I thought it meant if I specify strong1, for example. But I tried that in Verilog-XL and found that if I specify strong1, it really does get strong1. So the sentence is definitely unclear and possibly incorrect. -- Shalom Bresticker Shalom.Bresticker@motorola.com Design, Verification & Reuse Methodology Tel: +972 9 9522268 Motorola Semiconductor Israel, Ltd. Fax: +972 9 9522890 POB 2208, Herzlia 46120, ISRAEL Cell: +972 50 441478 |
| Fix |
In 7.8, CHANGE "The signals that these sources place on nets shall have pull strength in the absence of a strength specification. If conflicting strength specification is declared, it shall be ignored." TO "The signals that these sources place on nets shall have pull strength in the absence of a strength specification. If there is a strength1 specification on a pullup source or a strength0 specification on a pulldown source, the signals shall have the strength specified. A strength0 specification on a pullup source and a strength1 specification on a pulldown source shall be ignored." CHANGE the last sentence in 7.8 from "In this example, the p1 instance drives neta and the p2 instance drives netb." TO "In this example, the p1 instance drives neta and the p2 instance drives netb with strong strength." In 7.1.2, at the end of the paragraph discussing pullups and pulldowns, add "See 7.8 for more details." |
| Audit-Trail |
From: Steven Sharp <sharp@cadence.com> To: etf-bugs@boyd.com, Shalom.Bresticker@motorola.com Cc: Subject: Re: errata/526: 7.8 "conflicting strength is ignored" ?? Date: Tue, 6 Jan 2004 14:30:04 -0500 (EST) >What does "conflicting strength specification" mean ? I would guess that it means specifying a 0 drive strength on a pullup or a 1 drive strength on a pulldown. For example: pullup (weak0) p1(w); This will produce a warning in XL and NC, and drive w with pull1 (ignoring the weak0 specification). The warning in NC actually calls this a conflicting drive strength, which doesn't prove anything except that another implementor came to the same conclusion that I did about what this term was intended to mean. Steven Sharp sharp@cadence.com From: Shalom.Bresticker@motorola.com To: Steven Sharp <sharp@cadence.com> Cc: etf-bugs@boyd.com Subject: Re: errata/526: 7.8 "conflicting strength is ignored" ?? Date: Tue, 6 Jan 2004 22:39:08 +0200 (IST) But that is simply illegal, both according to the text (7.1.2) and the BNF (Syntax 7-1). So maybe it was intended to refer to the strength0 part of a (strength0, strength1) specification for a pullup (and opposite for pulldown)? Shalom On Tue, 6 Jan 2004, Steven Sharp wrote: > >What does "conflicting strength specification" mean ? > > I would guess that it means specifying a 0 drive strength on a pullup > or a 1 drive strength on a pulldown. For example: > > pullup (weak0) p1(w); > > This will produce a warning in XL and NC, and drive w with pull1 (ignoring > the weak0 specification). The warning in NC actually calls this a > conflicting drive strength, which doesn't prove anything except that > another implementor came to the same conclusion that I did about what > this term was intended to mean. -- Shalom Bresticker Shalom.Bresticker@motorola.com Design, Verification & Reuse Methodology Tel: +972 9 9522268 Motorola Semiconductor Israel, Ltd. Fax: +972 9 9522890 POB 2208, Herzlia 46120, ISRAEL Cell: +972 50 441478 From: Steven Sharp <sharp@cadence.com> To: sharp@cadence.com, Shalom.Bresticker@motorola.com Cc: etf-bugs@boyd.com Subject: Re: errata/526: 7.8 "conflicting strength is ignored" ?? Date: Tue, 6 Jan 2004 16:17:13 -0500 (EST) >But that is simply illegal, both according to the text (7.1.2) and the BNF >(Syntax 7-1). So maybe it was intended to refer to the strength0 part of a >(strength0, strength1) specification for a pullup (and opposite for >pulldown)? That makes sense. In that case, the extra strength is completely ignored, as the strange sentence says. Not matching the drive direction isn't really a conflict, but someone might have used that term anyway for this mismatching direction. This particular text does not appear in the original XL manual or the OVI 2.0 LRM, so it was presumably added during the IEEE standardization process. Maybe somebody from the 1364-1995 days can remember something. Steven Sharp sharp@cadence.com Fix replaced by Shalom.Bresticker@motorola.com on Thu Feb 12 07:23:52 2004 In 7.8, CHANGE "The signals that these sources place on nets shall have pull strength in the absence of a strength specification. If conflicting strength specification is declared, it shall be ignored." TO "The signals that these sources place on nets shall have pull strength in the absence of a strength specification. If there is a strength1 specification on a pullup source or a strength0 specification on a pulldown source, the signals shall have the strength specified. A strength0 specification on a pullup source and a strength1 specification on a pulldown source shall be ignored." CHANGE the last sentence in 7.8 from "In this example, the p1 instance drives neta and the p2 instance drives netb." TO "In this example, the p1 instance drives neta and the p2 instance drives netb with strong strength." In 7.1.2, at the end of the paragraph discussing pullups and pulldowns, add "See 7.8 for more details." |
| Unformatted |
|
Hosted by Boyd Technology