Subject: Re: [sv-ec] Asychronous mailbox proposal
From: Kevin Cameron x3251 (Kevin.Cameron@nsc.com)
Date: Fri Dec 13 2002 - 16:30:56 PST
> From: "Arturo Salz" <Arturo.Salz@synopsys.com>
>
> Message Stuart,
>
> It is possible to wait on dynamically allocated objects. The rule is that the handle
> is evaluated at the time at which the statement executes.
>
> The LRM still contains a $wait_var system task to wait for dynamic as well as static
> variables, but it is only there as part of the donation. I expect the @ operator to be
> able to handle dynamic objects as well. Some committee members have already
> suggested that.
>
> Arturo
If you are proposing we adopt $wait_var and roll it into @, I can't see why you are
objecting to @(<mailbox>).
For things other than nets, mailboxes and semaphores I suspect the simulation overhead
for $wait_var will be horrendous - is it actually implemented in any simulator?
$wait_var would in theory be user-replaceable too!
Kev.
>
> ----- Original Message -----
> From: Stuart Swan
> To: Arturo Salz ; Kevin Cameron ; sv-ec@eda.org
> Cc: david.smith@synopsys.COM ; Mehdi.Mohtashemi@synopsys.COM
> Sent: Friday, December 13, 2002 2:08 PM
> Subject: RE: [sv-ec] Asychronous mailbox proposal
>
>
> Arturo-
>
> I thought that it was only possible in SV to wait on objects that
> were not dynamically allocated. (Is this true?)
>
> In the example below, I assume that objects of class kmbox are dynamically
> allocated, and thus bit has_message is dynamically allocated, and thus
> you can't wait on it.. ?
>
> Stuart
...
This archive was generated by hypermail 2b28 : Fri Dec 13 2002 - 16:31:38 PST