Re: [sv-bc] Re: [sv-ec] task/function actuals for mode "ref"

From: Gordon Vreugdenhil <gordonv_at_.....>
Date: Tue Jan 15 2008 - 14:44:08 PST
I've uploaded a pdf version of the 2235 proposal.  I haven't
dealt with Shalom's other concerns until there is agreement
on language.

Gord.

Neil Korpusik wrote:
> Hi Gord,
> 
> I see the same duplication issues that Shalom mentioned. Here is
> what Firefox shows me for the "From" part of the proposal. I suspect
> it has to do with Microsoft html extensions that aren't supported by
> other browsers. The simplest way to work around this is to upload a pdf.
> 
>    "The semantics of assignments to variables passed by reference is 
> that changes
>     are seen outside the subroutine immediately (before the subroutine 
> returns). Only
>     variables, not nets, can be passed by reference.  Only variables, 
> not nets, can
>     be passed by reference."
> 
> Neil
> 
> 
> 
> Gordon Vreugdenhil wrote:
>>
>>
>> Bresticker, Shalom wrote:
>>> Gord,
>>>
>>> I don't think you mean 'ref mode', there is no such term as far as I
>>> know. I guess you mean 'ref arguments'.
>>>
>>> In the proposal, in the FROM text, you duplicated the sentence, "Only
>>> variables, not nets, can be passed by reference." It only appears once
>>> in the LRM. Then in the TO text, you struck out the second, but that
>>> still leaves the first.
>>
>> I am confused -- the second what?  I didn't duplicate that
>> sentence and I only found it once in the LRM.
>>
>>
>>> You use the term "indexed select of an unpacked array". The term
>>> "indexed select" does not appear in the LRM and is not defined. The
>>> closest is "indexed part-select", referring to the [n+:p] syntax. I
>>> don't think you meant that. Mantis 2169 also adds the term "non-indexed
>>> part-select" to refer to the [m:n] syntax.
>>
>> There really isn't a direct term for what I am trying to say.  The
>> closest is from 7.4.6:
>>    A single element of a packed or unpacked array can be selected 
>> using an
>>    indexed name.
>>
>> I meant an element of an unpacked array selected using a non-slice
>> indexed name.
>>
>> Please feel free to suggest a correct term, or I can use the
>> verbose wording above.
>>
>>> Your list of permitted ref arguments would seem to exclude
>>> concatenations. That would be a narrower interpretation than 2097's.
>>
>> Concats are not legal in a ref.  2097 in its entirety is not really
>> "narrower" or "wider"; the term "variable" is what was too
>> narrow for ref actuals.  There are some kinds of expressions
>> that are valid lvals that are NOT valid as a ref.  Concats
>> are one of those.  So 2097 allows some things that are not
>> legal in refs and refs allow some things that are not permitted
>> as the LHS of a force.
>>
>> Gord.
>>
>>

-- 
--------------------------------------------------------------------
Gordon Vreugdenhil                                503-685-0808
Model Technology (Mentor Graphics)                gordonv@model.com


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
Received on Tue Jan 15 14:44:30 2008

This archive was generated by hypermail 2.1.8 : Tue Jan 15 2008 - 14:45:06 PST