Verification Guild
A Community of Verification Professionals
Search


  Login  
Nickname

Password

Security Code:
Security Code
Type Security Code
BACKWARD
Don't have an account yet? You can create one. As a registered user you have some advantages like theme manager, comments configuration and post comments with your name.

  Modules  
  • Home
  • Downloads
  • FAQ
  • Feedback
  • Recommend Us
  • Web Links
  • Your Account

  •   Who's Online  
    There are currently, 116 guest(s) and 0 member(s) that are online.

    You are Anonymous user. You can register for free by clicking here

     
    Verification Guild :: View topic - e-Support in Synopsys' VCS ?
     Forum FAQForum FAQ   SearchSearch   UsergroupsUsergroups   ProfileProfile   Private MessagesPrivate Messages   Log inLog in 

    e-Support in Synopsys' VCS ?
    Goto page 1, 2  Next
     
    Post new topic   Reply to topic    Verification Guild Forum Index -> Main
    View previous topic :: View next topic  
    Author Message
    dieckmann
    Newbie
    Newbie


    Joined: Nov 03, 2007
    Posts: 1

    PostPosted: Mon Nov 05, 2007 9:52 am    Post subject: e-Support in Synopsys' VCS ? Reply with quote

    I heard from a colleague that Synopsys is claiming that they are supporting e natively within VCS. This is really great news since we believe e is the ideal verification language and with all of the talk these days about SystemVerilog, it is good to see that Cadence is the not the only vendor supporting e now.
    Has anyone had a chance to try out e support in VCS?
    Does anyone know how to get access to the version of VCS that supports e?
    I suspect that Synopsys is working closely with some early beta customers before they roll this out to the rest of us.
    Back to top
    View user's profile
    sarge
    Junior
    Junior


    Joined: Nov 05, 2007
    Posts: 6

    PostPosted: Mon Nov 05, 2007 1:47 pm    Post subject: Reply with quote

    I've heard the same thing. Since my company has recently decided to adopt Synopsys, I'm going to give some code I have lying around a whirl and see what happens. Don't know what to expect, but anything is better than nothing.
    Back to top
    View user's profile
    rader
    Newbie
    Newbie


    Joined: Oct 19, 2007
    Posts: 3

    PostPosted: Tue Nov 06, 2007 3:33 am    Post subject: e support in VCS Reply with quote

    Can anyone confirm if this rumor is true or false ?

    I would be very interested in finding out more details about exactly how much of "e" will be supported by Synopsys. I realize that e-code is now supposedly a "standard". But it would be great to really understand how much of our Cadence-based e-code environments would eventually be portable to Synopsys. Till now, with only Cadence supporting e-code, we have not had to worry too much about creating portable environments.

    I am concerned about things which might not necessarily be viewed as part of the e-code "standard". To be honest, I have not had much reason to understand which parts of e-code are in the "standard" and which are Cadence unique additions. Sequences leap to mind for one thing.
    Obviously existing Cadence eVC such as vr_ad would also present issues.
    But, if we can figure out the differences, we can start tailoring new environments to make the eventual porting to Synopsys e-code easier.

    Once we get e-code environments which can "easily" be ported to run on multiple vendors, we will no longer be strictly tied to one vendors flow vs another. Can you say "license negotiation leverage" ?
    Back to top
    View user's profile
    Darren
    Senior
    Senior


    Joined: Jan 06, 2004
    Posts: 32
    Location: Bristol, England

    PostPosted: Tue Nov 06, 2007 5:22 am    Post subject: Reply with quote

    Sequences are in the standard, or at least will be in the 2008 release. What won't be supported are calls to Specman specific features, such as trace facilities, or calls to other debug features. Most of the rest of it should be part of the standard language. Things such as vr_ad use the standard language so should be supported.
    Back to top
    View user's profile Send e-mail
    drchip
    Senior
    Senior


    Joined: Dec 10, 2003
    Posts: 14
    Location: Everywhere

    PostPosted: Tue Nov 06, 2007 9:22 am    Post subject: Reply with quote

    This is very interesting and potentially bad news for Mentor. I have heard a rumor that the next release of VCS will "support AVM 3.0". If VCS is also supporting e, this will allow only Cadence and Synopsys to fully support a mixed OVM/eRM environment - since most of eRM will probably be in OVM then Mentor will be left behind somewhat as Synopsys will be able to work with existing Cadence customers who've followed a URM flow. URM and OVM are probably more or less the same thing. Allowing customers to keep the e portions of their environment allows a less painful transition to OVM than would be the case. If questa, VCS and incisive all are capable of running AVM3.0 (and hence OVM) then questa will be at a disadvantage if it doesn't also support e, price and performance notwithstanding.

    I guess another question is what will happen to VMM? Presumably Synopsys will abandon this - if OVM is open source and runs on all three simulators, there's no compelling reason to lock-up an environment into a proprietary solution. I guess, effectively, URM aka OVM will become the dominant methodology.

    Dr Chip
    Back to top
    View user's profile
    Logger
    Senior
    Senior


    Joined: Jun 15, 2004
    Posts: 348
    Location: MN

    PostPosted: Tue Nov 06, 2007 12:42 pm    Post subject: Reply with quote

    drchip wrote:
    I guess another question is what will happen to VMM? Presumably Synopsys will abandon this - if OVM is open source and runs on all three simulators, there's no compelling reason to lock-up an environment into a proprietary solution.


    I don't ever see that happening. Synopsys has been adding support for features left and right, and I would see this as yet another feature. Meaning VCS will work with whatever methodology you choose.

    Janick is at Synopsys, and he is adding new layers of functionality on top of VMM to make it more powerful. I suspect VCS will support any methodology, but they will continue to promote and develop VMM if you ask them how to build your environment.

    The other simulators will gain support for VMM in response. And while Synopsys has not open sourced their implementation, there is already the TrustIC version that runs on Questa. Synopsys also does have a licensing program that let's you run their VMM on other tools (if they support it).

    I think we may be moving away from the language war and into the methodology war. Which means, as we saw with mixed-hdl, we'll be seeing solutions from the vendors which allow OVM and VMM components to work in the same environment.
    Back to top
    View user's profile
    Darren
    Senior
    Senior


    Joined: Jan 06, 2004
    Posts: 32
    Location: Bristol, England

    PostPosted: Wed Nov 07, 2007 11:28 am    Post subject: Reply with quote

    Anything which moves towards standard e-support amongst multiple vendors is good by me. It also makes it far easier when a vendor comes in trying to convince us to use new tool. Rather than us asking about legacy code and the cost of translating it all, we can concentrate on the things which really matter such as features, cost, speed etc etc. It's no good trying to sell me the fastest tool on the planet if it can't simulate what I want and it takes me a year to transcribe everything so it can (and I lose my market window slot in the process)!
    Back to top
    View user's profile Send e-mail
    akhan
    Newbie
    Newbie


    Joined: Jul 02, 2007
    Posts: 1

    PostPosted: Wed Nov 07, 2007 5:32 pm    Post subject: e-Support in Synopsys' VCS ? Reply with quote

    Interesting! What version of VCS supports e? Does it support eRM? It would be interesting to see how a mixed code of VMM+eRM reacts Smile
    Back to top
    View user's profile
    kev
    Senior
    Senior


    Joined: Aug 24, 2004
    Posts: 80

    PostPosted: Thu Nov 08, 2007 4:51 am    Post subject: Reply with quote

    this makes me angry Shocked

    Yet another way to do the same thing. I do not care for the language, just as long as there is a consensus over one.

    Methodologies are not language specific. Arguing over VMM, OVM, eRM, AVM features is healthy and beneficial.

    Arguing over the correct language to write them in and wasting time trying to get environments written in differing languages to talk to each other (with VIP adoption and legacy platforms) is neither.

    Synopsys adopting e just muddies the water again (if this is actually true)

    Kevin
    Back to top
    View user's profile Send e-mail
    sarge
    Junior
    Junior


    Joined: Nov 05, 2007
    Posts: 6

    PostPosted: Thu Nov 08, 2007 12:02 pm    Post subject: Reply with quote

    There is no one 'correct' language for everything or everyone, which is why it IS beneficial when languages can talk to each other. Everyone can then code in whichever language they prefer, or are most proficient in, and then mix them together.

    I believe Synopsys supporting/adopting e actually makes things clearer, in that one doesn't have to 'prove the superiority' of one language to the other now. You can use whichever one you feel more comfortable with to get the job done
    Back to top
    View user's profile
    rader
    Newbie
    Newbie


    Joined: Oct 19, 2007
    Posts: 3

    PostPosted: Thu Nov 08, 2007 7:47 pm    Post subject: Reply with quote

    I have to disagree about the point that the language itself doesnt matter.
    I agree you can do the same thing in multiple languages, but if one of those languages makes it very easy to do that task, while another language makes it more obscure, I will tend to favor the easier one.

    If all languages are the same, and it is only a matter of methodology,
    then why arent we all programming in assembly or hex.

    The methodology and the language are more tightly coupled than that.
    Sure any language will work, but will you hit the Time to Market window ?

    These "which language discussions" always reminds me of the quote in perl
    "A perl script is "correct" if it gets the job done before your boss fires you."

    So a verification language is "correct" if we can get the thing verified before the schedule slides too far to the right.

    personally I like more tools in the toolbox, and more the vendors supporting my toolbox, the better.
    Back to top
    View user's profile
    Mark
    Newbie
    Newbie


    Joined: Jan 06, 2004
    Posts: 4
    Location: Bristol, UK

    PostPosted: Thu Nov 15, 2007 7:04 am    Post subject: Reply with quote

    From what I have heard (and this could make sense) it would be a large company (let's say IBM Exclamation ) who could get Synopsys to support e.

    Maybe there is someone out there who can confirm this Smile ..... or maybe they won't Shocked.
    Back to top
    View user's profile Visit poster's website
    Boone
    Senior
    Senior


    Joined: Jan 12, 2004
    Posts: 32

    PostPosted: Tue Nov 27, 2007 11:29 pm    Post subject: Reply with quote

    Synopsys signed a deal with Intel several months back, and Intel is a large e customer, so...

    I still think it's silly to revive e or VERA. If we can all just focus on one testbench language, then perhaps the EDA vendors can focus their energies on a complete and quality implementation of one language.
    Back to top
    View user's profile
    Darren
    Senior
    Senior


    Joined: Jan 06, 2004
    Posts: 32
    Location: Bristol, England

    PostPosted: Wed Nov 28, 2007 3:48 am    Post subject: Reply with quote

    It's not reviving e - it's recognising what is already there. Those of us who have many years investment in e infrastructure are not going to throw it all away and rewrite the lot SystemVerilog just because it is the newest way to do things - we have to be able to run and support legacy systems. It has already cost a lot of time and effort to write and debug the e-code, and we don't want to incur the same cost (or more) to do the same thing in SV just to get back to where we are today. By all means focus on getting one quality testbench language which we can all use, but I'm not so sure that SV is it yet...... Especially if most of your RTL is in VHDL and you need the added cost of dual-language simulators to run SV (so there goes the cost advantage of using "one" simulator to do both testbench and design simulation).

    Note from the moderator: this topic was split to this topic.
    Back to top
    View user's profile Send e-mail
    sarge
    Junior
    Junior


    Joined: Nov 05, 2007
    Posts: 6

    PostPosted: Thu Nov 29, 2007 2:57 pm    Post subject: Reply with quote

    Yeah, how dare we have more than one testbench language lying around! It's not like we have more than one RTL design language lying around....oops, or more than one programming language....oops, or more than one....awwww forget it.
    Back to top
    View user's profile
    Display posts from previous:   
    Post new topic   Reply to topic    Verification Guild Forum Index -> Main All times are GMT - 5 Hours
    Goto page 1, 2  Next
    Page 1 of 2

     
    Jump to:  
    You cannot post new topics in this forum
    You cannot reply to topics in this forum
    You cannot edit your posts in this forum
    You cannot delete your posts in this forum
    You cannot vote in polls in this forum

    Powered by phpBB © 2001, 2005 phpBB Group
    Verification Guild (c) 2006-2014 Janick Bergeron
    PHP-Nuke Copyright © 2005 by Francisco Burzi. This is free software, and you may redistribute it under the GPL. PHP-Nuke comes with absolutely no warranty, for details, see the license.
    Page Generation: 0.12 Seconds