Hi, Not sure about your first question. All I know is that keeping track of every cache line will seriously hurt performance.
We support both MESI and MOESI, we already have a machine setup, moesi_private_L2. With mesi configurations in default.conf Brendan On Mon, Mar 25, 2013 at 7:07 PM, Zoran Jaksic <[email protected]> wrote: > Hello to everyone, > > First of all I would like to thank you for making this kind of simulator. > It seems like a really useful tool that I am planing to use for my future > study. > > I wanted to ask 2 questions if you can help me. > 1) My research includes possible implementation of dynamic memories in L1 > and L2 cache but in order to simulate that I'll need somehow to incorporate > "retention time" parameter (Time after the writing the world it becomes > invalid and it needs to be refreshed) in cache simulation. Basically what I > need is this: after writing the world in the cache I have to count number > of cycles and after certain time threshold to react. Can you advice me > somehow where it would be the best to do that. I was thinking to change > specific cache protocols file but is it going to be enough if I change > that. > > 2) I also wanted to ask since I didn't understood which cache coherency > protocol is implemented. is it MESI or MOESI or both? I was confused a > little bit but I think that on the website you mention that only MESI is > implemented on one place and on other place you mention MOESI. > > I would really like to thank you in forward and I hope that you continue > to support the MARSS in the future. > > best regards > Zoran > > > > -- > Zoran Jakšić > ------------------------------------------------------ > Research Assistant > Computer Architecture Department > Universitat Politecnica de Catalunya > ------------------------------------------------------- > Address: Campus Nord UPC C6-E208 > CR. Jordi Girona, 1-3 > 08034 Barcelona, Spain > ------------------------------------------------------- > Tel: +34648025808 > e-mail: [email protected] > [email protected] > <[email protected]> > ------------------------------------------------------- > > > _______________________________________________ > http://www.marss86.org > Marss86-Devel mailing list > [email protected] > https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel > >
_______________________________________________ http://www.marss86.org Marss86-Devel mailing list [email protected] https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
