Timothy M Jones wrote: > On Tue, 19 Oct 2010 22:31:25 -0400, Gabe Black <[email protected]> > wrote: > >> So if translation finishes immediately, the fault gets set in >> finishTranslation iniline with the initiateTranslation call, and when >> you get back to readBytes/writeByte you can check it. If not, the >> translation just cleans up after itself and disappears, and the fault is >> lost. >> >> Am I interpreting this correctly? Is this broken, or am I just missing >> some element that makes it work? Making this work correctly would make >> implementing my initiateAcc commits change easier because then the >> faults from executing initiateAcc and translation are clearly separated, >> not mushed together like they are now. >> > From what I remember when I was implementing the split translations in > O3, what you're saying is correct, Gabe. I had noticed this, but > never found a case where translation didn't finish immediately, so > didn't worry too much about it at the time. > > Tim >
Ok, thanks for clarifying. I think for the combination of ISAs and modes that work in O3 as of today, you were right that they don't use delayed translation. Gabe _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
