[sv-ac] RE: Mantis 2476 ($onehot, etc, outside assertions)

resend

From: Bresticker, Shalom
Sent: Tuesday, August 31, 2010 8:41 PM
To: Seligman, Erik; sv-ac@eda.org
Subject: RE: Mantis 2476 ($onehot, etc, outside assertions)

$countones is a "simple boolean function", not a "sampled value function".

At least regarding the sampled value functions, 16.9.3 says "The use of these functions is not limited to assertion features; they may be used as expressions in procedural code as well."

Regarding the global clocking functions, 16.9.4 says, "The global clocking sampled value functions may be invoked only in property_expr or in sequence_expr;
this implies that they shall not be used in assertion action blocks."

So, the only ambiguity today is about the "simple booleans".

Shalom

From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On Behalf Of Seligman, Erik
Sent: Tuesday, August 31, 2010 6:45 PM
To: sv-ac@eda.org
Subject: [sv-ac] Mantis 2476 ($onehot, etc, outside assertions)

Hi guys-I just added this note at http://www.verilog.org/mantis/view.php?id=2476 . Perhaps we can discuss in today's meeting.

We need to decide not just for the functions in the original description, but for all the assertion system functions:
$onehot $onehot0 $isunknown $sampled $rose $fell $stable $changed $past $countones $past_gclk $rose_gclk $fell_gclk $stable_gclk $changed_gclk $future_gclk $rising_gclk $falling_gclk $steady_gclk $changing_gclk

I think the return types are clear in the final 2009 LRM, but the question of their usage outside assertions is left ambiguous. For each of these categories, do we want to enable outside assertions? My instinct is just to enable category #1.
1. Simple booleans ($onehot $onehot0 $isunknown)
2. Sampled value functions ($sampled $rose $fell $stable $changed $past $countones)
3. Global clocking past functions ($past_gclk $rose_gclk $fell_gclk $stable_gclk $changed_gclk)
4. Global clocking future functions ($future_gclk $rising_gclk $falling_gclk $steady_gclk $changing_gclk)

Interestingly, 20.13 seems to have an isolated BNF production "assert_boolean_function" that is not referenced anywhere else in the LRM, leaving it unclear where this can be used, though the implication is that it's restricted to assertions. Should we fix this & Annex A to properly integrate these functions into the grammar as well?

--
This message has been scanned for viruses and
dangerous content by MailScanner<http://www.mailscanner.info/>, and is
believed to be clean.
---------------------------------------------------------------------
Intel Israel (74) Limited
This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.