Hi Hamid,
I think the mapping is done something like this:
ruby_proc_id (within a chip) = simics_proc_no/g_PROCS_PER_CHIP;
//(integer division)
simics_proc_no is in [0,7] and
Hence, first four simics-processors map to four proc in first chip
and next four simics-processor map to four procs in in the second
chip in ruby.
Regards,
Venkat
On 05/11/2011 12:57 AM, hamid reza khaleghzadeh wrote:
Hello
Suppose that g_NUM_PROCESSORS is 8 and g_PROCS_PER_CHIP is 4.
Therefore, there is 2 chip that eache of them contain 4 physical
processors. On the other hand, each physical processor in Simics
has a number (Suppose 8 cpus are defined in Simics). Could you
tell me how I can determine which processors belong to chip1? On
the other words, Which processors are shared one L2 cache?
Best Regards.
--
Hamid Reza Khaleghzadeh
_______________________________________________
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.
--
Venkatanathan Varadarajan,
Graduate Student,
Department of Computer Sciences,
University of Wisconsin - Madison.
http://pages.cs.wisc.edu/~venkatv/
|
|