IBIS Macromodel Task Group Meeting date: 15 October 2013 Members (asterisk for those attending): Agilent: * Fangyi Rao * Radek Biernacki Altera: * David Banas Julia Liu Hazlina Ramly Andrew Joy Consulting: Andy Joy ANSYS: Samuel Mertens * Dan Dvorscak * Curtis Clark Steve Pytel Luis Armenta Arrow Electronics: Ian Dodd Cadence Design Systems: Terry Jernberg * Ambrish Varma Feras Al-Hawari * Brad Brim Kumar Keshavan Ken Willis Cavium Networks: Johann Nittmann Celsionix: Kellee Crisafulli Cisco Systems: Ashwin Vasudevan Syed Huq Ericsson: Anders Ekholm IBM: Greg Edlund Intel: Michael Mirmak Maxim Integrated Products: Mahbubul Bari Hassan Rafat Ron Olisar Mentor Graphics: * John Angulo Zhen Mu * Arpad Muranyi Vladimir Dmitriev-Zdorov Micron Technology: * Randy Wolff * Justin Butterfield NetLogic Microsystems: Ryan Couts Nokia-Siemens Networks: Eckhard Lenski QLogic Corp. James Zhou SiSoft: * Walter Katz Todd Westerhoff Doug Burns * Mike LaBonte Snowbush IP: Marcus Van Ierssel ST Micro: Syed Sadeghi Teraspeed Consulting Group: Scott McMorrow * Bob Ross TI: Casey Morrison Alfred Chong Vitesse Semiconductor: Eric Sweetman Xilinx: Mustansir Fanaswalla Ray Anderson The meeting was led by Arpad Muranyi ------------------------------------------------------------------------ Opens: - None -------------------------- Call for patent disclosure: - None ------------- Review of ARs: - Walter create EMD vs. BIRD 145 example - Done ------------- New Discussion: Walter showed a BIRD 145/160 IBIS example - Walter: A basic difference is the use of MODELCALL instead of model name. - The assumption of one-to-one mapping of pin to die pad is not good for power networks - In the Ports section Pin_name is the same as IBIS Pin Number. - Arpad: What does Pin_name mean in Buffer entries? - Walter: It denotes a buffer instance. - Radek: Is Terminal syntactically required in the Buffer leaf? - Walter: Yes. - Radek: Will the Pad leafs be extended someday? - Walter: Other data will be needed when there is not one-to-one correspondence for power. - Then it will reference die pad names instead of pin names. - Walter showed what a [Die Pad] section might look like. - Walter: With BIRD 145 a DQ [Model Call] must be created. - Parsing and lookups are required to see what a pin is connected to. - Arpad: The [Node Declarations] also provide names. - Walter: The syntax is more spread out with BIRD 145. - With a memory controller at least 130 lines would be needed. - Arpad: The EMD like example would have at least 96 extra lines. - Also it is shorter only because it has no on-die nodes. - Radek: Walter's solution seems more compact. - What is the significance of the numbered leafs? - Walter: They are the subckt terminal position. - They could have been left off and line order assumed to be terminal order. - But explicit numbering allows using only some terminals. - John: How will we handle one-to-multiple mapping of pin to buffer? - Walter described a way to handle stacked memory and agreed this should be resolved. - A pin.instance format could be used. - John: A Stack leaf could be added instead. - Walter: We just need some kind of handle to identify which buffer we are talking about. - This is like EMD, with names for internal instances. - John: It would be good to avoid the BIRD 145 syntax. - But, it may solve the multiple instance problem already. - Walter: We have to ask which is easier for EDA tool parsing and debugging. - Bob: Which EMD Like pin is the pullup reference? - Walter: The one with Pu_Ref. - It is the internal pullup reference of the IBIS model. - Arpad: The port numbers are defined by the subckt. - Brad: How does this interact with the IBIS file? - Walter: Somewhere in the IBIS there would be a keyword to introduce the on-die IBIS-ISS file. - Also there would be one to introduce the package IBIS-ISS file. - Radek: We could look at what BIRD 125 has for that. - Calling it EMD probably adds confusion. - Walter: We have to be careful to say "EMD Like" - Brad: EMD instantiates the IBIS model. - I thought that would be proposed here. - As it stands this does not change the point of control. - Brad: What if I have a separate RDL? - Can the package and on-die interconnect be grouped? - Walter: I would like to see an example of that. - Randy: RDL is on-die - Arpad: Can circuits be cascaded? - Brad: It can be - Brad: If I have interposers and other connects does that need EMD? - Walter: Yes - John: It may be wise to look at the more involved cases now. - BIRD 145 seems more general, for example no one-to-one assumption - Walter: That does assume one-to-one. - We should look at more examples. - I can create examples, but others need to explain what is needed. AR: Arpad request people to create ideas for new interconnect examples ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives