Stephan Diestelhorst wrote on Friday 08 February 2013, 15:05:52:
> It seems that Marss (as opposed to PTLsim) is using the virtual
> addresses, and also does not do word (64 bit) aligned accesses
> anymore, but instead accesses sub-words at natural alignment.

Good lord.  I just continued digging through this and found that memory
accesses are actually *not aligned* at all.  That explains some of the
strange logic when computing store-to-load forwarding (which I am
currently redoing )-:

Any clue why that logic was ripped out?  Too complicated?

Stephan



_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel

Reply via email to