Re: [Gems-users] Opal "Caught time breakpoint"

[Luke Yen <lyen@xxxxxxxxxxx>]

  If you feel like modifying Opal code, you can place a
SIM_break_simulation() in opal/system/pseq.C's retireInstruction()
function.  You can count how many instructions (for any processor) has
retired by tracking how many times d->Retire() gets called in that
function.

   Luke

On Wed, 16 May 2007, Lei Yang wrote:

> Could anybody please shine some light on this? Do I need to implement a hap on breakpoints in this situation?
>
> Thanks!
>   ----- Original Message -----
>   From: Lei Yang
>   To: Gems-users@xxxxxxxxxxx
>   Sent: Wednesday, May 16, 2007 5:40 PM
>   Subject: [Gems-users] Opal "Caught time breakpoint"
>
>
>   Dear list,
>
>   When opal is used, what do I need to do to stop the simulation when a certain CPU has executed a certain number of instructions? For example, I have a 2 core CMP simulation, where I want CPU0 to execute exactly 10000 instructions. Without Opal, I can simply say "cpu0.step-break 10000", and the simulation will stop when cpu0 has executed 10000 instructions. But when opal is installed, opal starts simics with a large number of steps like "opal0.sim-step 10000000000" (documented in quickstart). In this case, when cpu0 has executed 10000 instructions, in the command line this will appear "Caught time breakpoint", but the simulation doesn't stop!? How do I work around this?
>
>   Thanks for your comments!
>
>   Lei
>
>
> ------------------------------------------------------------------------------
>
>
>   _______________________________________________
>   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.
>
>