Re: Summary of minor issues

John-san,

I understand that SystemC has never included ports for access to
mutexes and semaphores. We agree with your proposal of sc_mutex and
sc_semaphore.

Including example codes using sc_mutex or sc_semaphore can make clear
how we should use them. Also, the backward incompatibility should be included in the LRM.

Best regards,
Hiroshi Imai
Chair of SystemC WG, JEITA

At 09 Apr 2010 13:28:10 +0100 john.aynsley@doulos.com wrote:
> Hiroshi-san,
>
> I'm not sure I understand your point. The SystemC standard has never
> included ports for access to mutexes and semaphores, and this proposed
> change does not affect that. The only change is that the base class of
> sc_mutex/semaphore themselves will become sc_object instead of
> sc_prim_channel, which gives more flexibility when instantiating these
> objects. This change would NOT be backward-compatible if any existing code
> exploits the fact that they are primitive channels (by using the
> request-update mechanism), but that seems unlikely.
>
> Regards,
>
> John A
>
>
>
>
> From:
> Hiroshi Imai <hiroshi3.imai@toshiba.co.jp>
> To:
> john.aynsley@doulos.com
> Cc:
> systemc-p1666-technical@eda.org
> Date:
> 08/04/2010 03:05
> Subject:
> Re: Summary of minor issues
>
>
>
> [snip]
>
> In our understanding, this change makes modules enable to access
> sc_mutex and sc_semaphor without through sc_port. Is it correct?
> We are wondering that this change may have negative influence on
> interoperability of user models. Any comments?
>
> [snip]
>
> Best regards,
> Hiroshi Imai
> Chair of SystemC WG, JEITA
>
>
>
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
>

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.