[great-users] Undefined Semantics

[Attila Vizhanyo]

Abdullah,

The semantics of GR packets are the most particular in the language,
and can lead to the confusions you seem to have.

If packets are expressed in terms of pattern objects, one packet
consists of a number of objects, each object directed to a GR port.
Now, any rule can have arbitrary number of ports, and each port is
supplied with objects from (possibly various) sources. Take one object 
out of each port, and you get a GR packet. This simple rule has
the following implications
1, a packet is formed out of objects coming from ALL input/ output ports
of a given rule
2, the number of objects must be equal at ALL inputs/ output ports of a
given rule
3, if one source rule supply objects for an input port of a destination
rule, it must supply objects for all input ports of that destination.

Please refer to the GReAT User Manual for more info.

-Attila

-----Original Message-----
From: Abdullah Sowayan 
Sent: Saturday, January 03, 2004 21:18
To: great-users at list.isis.vanderbilt.edu
Subject: [great-users] Undefined Semantics

Hi Ya'll

What exactly are the semantics of a Block? Say I have 4 inputs ports for
a block. The first 2 input ports of the block are connected to Rule1,
the second 2 input ports are connected to Rule2. If a packet arrives at
the first 2 input ports, but nothing arrives at the second 2 inputs
ports, will these packets be pushed to rule1 or will it have to wait
until are input ports receive a packet?

How about output ports of a block? A similar question to the one above
also follows.

Thanks,
Abdul

-----Original Message-----
From: Abdullah Sowayan 
Sent: Friday, January 02, 2004 6:46 PM
To: great-users at list.isis.vanderbilt.edu
Subject: [great-users] GRD Debugger

Hi Ya'll

Earlier today there was a problem GRE Engine with respect to cross links
(didn't support cross links for more than 2 meta-models). Thanks to
Feng, that got fixed.

I could be wrong, but it seems to me that the GRD (the debugger) has the
same limitation with cross links as mentioned above (it is giving the
same error at exactly the same location with respect to cross links).

I'll be glad to demonstrate this to whom ever is interested in this.

Take care,
Abdul
_______________________________________________
great-users mailing list
great-users at list.isis.vanderbilt.edu
http://list.isis.vanderbilt.edu/mailman/listinfo/great-users
_______________________________________________
great-users mailing list
great-users at list.isis.vanderbilt.edu
http://list.isis.vanderbilt.edu/mailman/listinfo/great-users