( ESNUG 461 Item 13 ) ------------------------------------------- [01/31/07]

Subject: ( ESNUG 456 #10 ) We're still using the old Palladium emulator

> Palladium II's biggest drawbacks...
>
> I have the same comments for Palladium II as I did for Palladium I: it
> is a very complex, large environment and very expensive, so it could
> always be smaller and cheaper.  However, this is an area where nothing
> else but speed matters, which helps to justify the expense.
>
>     - Joerg Kayser
>       IBM Labs                                   Boeblingen, Germany


From: [ Doctor Whom ]

Hi, John,

We picked the Palladium emulator for its acceleration and because it can
support several users simultaneously.  I personally started using Palladium
in July 2005.  The 2 main ways we use it are:

  1. Targeted firmware verification.  Our environment is set up to verify
     firmware using our hardware design models.

  2. Duplicating test cases for customer support.  If a customer calls
     us with issues in their SW using our chips, we can find the problem
     by reconstructing the situation in Palladium.  For example, a customer
     called our engineers with video synchronization issues on one of our
     designs.  We were able to quickly duplicate the set-up in Palladium
     and resolve the problem.

We have our own set of targeted firmware and hardware tests, and have an
automated regression flow.  This is run every time there is a rev change in
our hardware or software models.

Palladium's upsides are:

  - Multi-tasking.  Both our software and hardware designers can run
    Palladium simultaneously in multiple domains.

  - Speed.  Our design can be synthesized and compiled into Palladium,
    on a Solaris platform, in just under 15 minutes.  Using the latest
    Palladium software, and running on a Linux based platform, we've
    found synthesis and compile time can be reduced to 3 to 4 minutes.

  - Palladium supports interactive debugging.  We use its full vision
    option in Novas Debussy.

Palladium's downsides are:

  - It's SVA support is untested.  We are looking at assertion-based sims
    in the Palladium environment, and are considering using System Verilog
    assertions that are supposedly available with Palladium II.

  - Palladium is a real headache to setup.  But once Cadence's support
    engineers set it up, it has been relatively straightforward to use.

  - Cost.  Palladium isn't cheap.  If your management isn't seriously
    committed to emulation, don't bother calling the Cadence sales guy.

  - I headed the effort of adding additional software debug capabilities
    to our Palladium system.  This involved adding JTAG and ETM support
    via Green Hills' debugger environment and their Super Trace probe
    hardware.  This task was long, difficult, and hampered by the fact
    that now a third party was in the mix. Getting support from both
    companies to debug something was not easy.  In the end, we now have
    this debug capability, but it took a substantial commitment in time
    and resources to get this to work.

Overall Palladium allowed us to do real world tests on our design before
silicon.  I would recommend it.  It will be interesting to see how their
new Palladium II works out.

Please keep me anon if you publish this, John.

    - [ Doctor Whom ]
Index    Next->Item







   
 Sign up for the DeepChip newsletter.
Email
 Read what EDA tool users really think.


Feedback About Wiretaps ESNUGs SIGN UP! Downloads Trip Reports Advertise

"Relax. This is a discussion. Anything said here is just one engineer's opinion. Email in your dissenting letter and it'll be published, too."
This Web Site Is Modified Every 2-3 Days
Copyright 1991-2024 John Cooley.  All Rights Reserved.
| Contact John Cooley | Webmaster | Legal | Feedback Form |

   !!!     "It's not a BUG,
  /o o\  /  it's a FEATURE!"
 (  >  )
  \ - / 
  _] [_     (jcooley 1991)