Thanks again Ali.
'overall_mshr_misses' in the performance model is exactly what I need to
compare with 'overal_misses' in the functional model.
Now I have pretty good correlation between the models:
L1I-misses: 8369552 (4.9 / 1k) / 9574348 (5.6 / 1k) (+14%)
L1D-misses: 10526355 (6.2 / 1k) / 10837326 (6.4 / 1k) (+3%)
L2(I)-misses: 516328 (0.3 / 1k) / 623182 (0.4 / 1k) (+21%)
L2(D)-misses: 3610866 (2.1 / 1k) / 3622024 (2.1 / 1k) (+0.3%)
Many thanks!
Orangeade
> Hi Orangeade,
>
> What about the misses per thousand instructions (E.g. device MSHR misses by
> instructions) for both models?
>
> Ali
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
