get_option takes a bool reference and will effectively set that flag to
true based on the YAML config.

https://github.com/avadhpatel/marss/blob/5b144b46cec2c01939f3f2fbf13ecd4d719bc11a/ptlsim/sim/machine.h#L104

https://github.com/avadhpatel/marss/blob/5b144b46cec2c01939f3f2fbf13ecd4d719bc11a/ptlsim/sim/machine.cpp#L476

Tyler

> Dear marssx86 developers:
>
> I appreciate your beautiful simulator.
> However, I'm afraid if it currently has simple but critical buggy codes,
> regarding setting up mesiLogic and moesiLogic.
>
> The associated files (ptlsim/cache/mesiLogic.cpp and
> ptlsim/cache/moesiLogic.cpp) forcibly has both variables of
>      is_private
>      is_lowest_private
> set to false, and accordingly, will always call both setup functions of
>      cont->set_private(is_private);
>      cont->set_lowest_private(is_lowest_private);
> with false boolean value.
>
>
> FYI, here I put the corresponding codes.
> 1. In ptlsim/cache/mesiLogic.cpp:
> 476         bool is_private = false;
> 477         if (!mem.get_machine().get_option(name, "private",
> is_private)) {
> 478             is_private = false;
> 479         }
> 480         cont->set_private(is_private);
> 481
> 482         bool is_lowest_private = false;
> 483         if (!mem.get_machine().get_option(name, "last_private",
> 484                     is_lowest_private)) {
> 485             is_lowest_private = false;
> 486         }
> 487         cont->set_lowest_private(is_lowest_private);
>
> 2. In ptlsim/cache/moesiLogic.cpp:
> 498         bool is_private = false;
> 499         if (!mem.get_machine().get_option(name, "private",
> is_private)) {
> 500             is_private = false;
> 501         }
> 502         cont->set_private(is_private);
> 503
> 504         bool is_lowest_private = false;
> 505         if (!mem.get_machine().get_option(name, "last_private",
> 506                     is_lowest_private)) {
> 507             is_lowest_private = false;
> 508         }
> 509         cont->set_lowest_private(is_lowest_private);
>
>
> Please correct me if I misunderstand.
>
> Thank you.
>
>
>
> Sincerely,
> Suk Chan Kang
>
>
> _______________________________________________
> 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

Reply via email to