On Wed, 19 Oct 2011, Korey Sewell wrote:
Nilay,
I was thinking you can run some splash benchmarks (the config file is
already setup configs/example/splash2?) and then catch the situation in the
classic memory system where an invalidate is forwarded to the CPU. Then, for
Ruby you can make sure that the same interface is being maintained for the
same situation.
Does that make sense?
On Wed, Oct 19, 2011 at 6:22 PM, Nilay Vaish <[email protected]> wrote:
Hi
I had recently put up a patch on the review board for enabling Ruby to work
with the O3 CPU. I have not been able to test so far. My understanding is
that pthreads based don't work with gem5 currently. Then, how is the O3 CPU
tested with the classic memory system? May be we can use the same tests in
case of Ruby.
Thanks
Nilay
Korey, the timings for the two memory systems can be different. I was
thinking if we can develop small programs based on published tests.
Secondly, how do you support threads in SE?
Thanks
Nilay
_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev
