Re: [Gems-users] Wattch SIGSEGV on Opal init

[Luke Yen <lyen@xxxxxxxxxxx>]

Hi Fred,

  Can you narrow down the cause of the segfault?  You can attach gdb to 
the simulation process, or debug by adding printfs.
  There's no wiki documentation on how to use WATTCH with Opal.  Most of 
the customizable parameters are in power.h, and the power measurements 
require an understanding of how WATTCH models the power of structures. 
Take a look at the original WATTCH paper for details.  It might also help 
if you download the original WATTCH package for SimpleScalar.
  Luke


On Thu, 28 Aug 2008, Fred A Bower wrote:

> I have an interest in using the integrated wattch feature on Gems 2.1 with
> Simics 3.0.x (I've got both 3.0.29 and 3.0.31 installed in a couple of
> forks of my simulation environment tree).
>
> I've built with gcc 3.4.6 on both 32-bit Intel and 64-bit AMD Linux
> environments. When I set the WATTCH_POWER value to 1, Opal causes a
> SIGSEGV on the call to init. Combing the archives, I found an instance of
> another user with the same problem, but no posted resolution (
> https://lists.cs.wisc.edu/archive/gems-users/2008-January/msg00083.shtml
> ). Further, the notes (
> https://lists.cs.wisc.edu/archive/gems-users/2008-February/msg00138.shtml
> ) on wattch simply indicate that there's no documentation and to set
> WATTCH_POWER to 1. Any pointers toward getting some power measurements are
> welcome.
>
> Thanks,
> fred
>
> Department of Computer Science, Duke University, Durham, NC 27708-0129
> Email:  fredb
> _______________________________________________
> Gems-users mailing list
> Gems-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/"; to your search.