Now, it's not very consistent in the code, and the stats have to be
fixed in one way or another.
Lisa Tue Feb 23 09:34:22 2010 -0800: addrRange(p->addr_range),
Lisa Tue Feb 23 09:34:22 2010 -0800: #if FULL_SYSTEM
Lisa Tue Feb 23 09:34:22 2010 -0800: .init(_numCpus + 1)
Lisa Tue Feb 23 09:34:22 2010 -0800: #else
Lisa Tue Feb 23 09:34:22 2010 -0800: .init(_numCpus)
Lisa Tue Feb 23 09:34:22 2010 -0800: #endif
Lisa Tue Feb 23 09:34:22 2010 -0800: #if FULL_SYSTEM
Lisa Tue Feb 23 09:34:22 2010 -0800: .init(_numCpus + 1)
Lisa Tue Feb 23 09:34:22 2010 -0800: #else
Lisa Tue Feb 23 09:34:22 2010 -0800: .init(_numCpus)
Lisa Tue Feb 23 09:34:22 2010 -0800: #endif
rdreslin Wed Jun 28 11:02:14 2006 -0400: .init(maxThreadsPerCPU)
stever Sun Jun 17 17:27:53 2007 -0700: .init(maxThreadsPerCPU)
stever Sun Jun 17 17:27:53 2007 -0700: .init(maxThreadsPerCPU)
stever Sun Jun 17 17:27:53 2007 -0700: .init(maxThreadsPerCPU)
stever Sun Jun 17 17:27:53 2007 -0700: .init(maxThreadsPerCPU)
stever Sun Jun 17 17:27:53 2007 -0700: .init(maxThreadsPerCPU)
stever Sun Jun 17 17:27:53 2007 -0700: .init(maxThreadsPerCPU)
stever Sun Jun 17 17:27:53 2007 -0700: .init(maxThreadsPerCPU)
stever Sun Jun 17 17:27:53 2007 -0700: .init(maxThreadsPerCPU)
as in src/mem/cache. Should I change the stever style to Lisa style?
On Thu, Oct 20, 2011 at 9:15 PM, nathan binkert <[email protected]> wrote:
>>> Looks like it's Lisa's fault ;-)
>>>
>>> http://repo.m5sim.org/m5/diff/ab05e20dc4a7/src/mem/cache/base.cc
>>>
>>> I think Nate's point is that all the stats vector lengths should be
>>> changed to _numCpus or _numCpus+1 instead of maxThreadsPerCpu to be
>>> consistent.
>>>
>>> We should also either (1) always do _numCpus+1 even though the extra
>>> "device" slot is unnecessary for SE mode
>>
>> What does device mean here? I'm not sure why there's a "+1"?
>
> In the end, this isn't the route that we're going, but basically,
> we're trying to keep track of cache stats on a per cpu basis, but in
> reality, there are things other than cpus that initiate cache
> requests. Those things (under this proposal) are binned in that +1.
>
> Nate
> _______________________________________________
> gem5-dev mailing list
> [email protected]
> http://m5sim.org/mailman/listinfo/gem5-dev
>
_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev
