Re: [sv-bc] RE: Connecting generated interface instances

From: Rich, Dave <Dave_Rich@mentor.com>
Date: Tue Mar 03 2015 - 12:50:50 PST
Recursion might be an extreme example. I think there is a desire to keep the elaboration process  close to a top-down one-pass flow.

- Dave
Sent from Outlook<http://taps.io/outlookmobile>




On Tue, Feb 24, 2015 at 10:45 AM -0800, "Bresticker, Shalom" <shalom.bresticker@intel.com<mailto:shalom.bresticker@intel.com>> wrote:

And if we removed the restriction, would it be a disaster?

One can always do things within the standard that cannot be handled.
For example, a recursion that never terminates.

From: Mark Hartoog [mailto:Mark.Hartoog@synopsys.com]
Sent: Tuesday, February 24, 2015 16:58
To: Bresticker, Shalom; Mark Hartoog; Steven Sharp; sv-bc@eda.org
Subject: RE: Connecting generated interface instances

The LRM clearly allows you to access named typedefs through the interface port.  Those named typedefs can depend on parameters of the interface instance. If we remove the restriction on a hierarchical reference, then we would have to remove the named typedef through interface ports or put a bunch of restrictions on the use of those types that would be very hard to describe or understand.

This would not be backwards compatible with the current language.

From: Bresticker, Shalom [mailto:shalom.bresticker@intel.com]
Sent: Tuesday, February 24, 2015 6:53 AM
To: Mark Hartoog; Steven Sharp; sv-bc@eda.org<mailto:sv-bc@eda.org>
Subject: RE: Connecting generated interface instances

Or the reverse ?

From: Mark Hartoog [mailto:Mark.Hartoog@synopsys.com]
Sent: Tuesday, February 24, 2015 16:51
To: Bresticker, Shalom; Mark Hartoog; Steven Sharp; sv-bc@eda.org<mailto:sv-bc@eda.org>
Subject: RE: Connecting generated interface instances

It is not a problem as long as we keep the restriction that the high conn of the interface port is not a hierarchical reference through generate blocks. In this case it is similar to a defparam.

We probably only give an error if it is a hierarchical reference through a generate block that has not yet been unrolled, but that gets complicated to specify and no two tools can be expected to unroll generate blocks in the same order.

The only simple, understandable restriction is a high conn hierarchical reference cannot be through generate blocks.


From: Bresticker, Shalom [mailto:shalom.bresticker@intel.com]
Sent: Tuesday, February 24, 2015 6:41 AM
To: Mark Hartoog; Steven Sharp; sv-bc@eda.org<mailto:sv-bc@eda.org>
Subject: RE: Connecting generated interface instances

Which seems to indicate that it is not really such a problem…

I think we gave up trying to insist it was illegal some years ago because you already can legally access the same information other ways.

---------------------------------------------------------------------
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.

---------------------------------------------------------------------
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.

---------------------------------------------------------------------
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<http://www.mailscanner.info/>, and is
believed to be clean.

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
Received on Tue Mar 3 12:51:13 2015

This archive was generated by hypermail 2.1.8 : Tue Mar 03 2015 - 12:51:22 PST