You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi sir:
I found that the when a load is scheduled because of it having partitial address overlapped with a older store,
it will be resshduled by IEW::rescheduleMemInst --> InstructionQueue::rescheduleMemInst where eventully set the translationCompleted flag and translationStated flag to false
then the load when be issue again will have to do address translation one more time while it has done the translation completely
so why should be this flow? is it be better that just re-do the data request by the physical address having translated before ?
or is this a hardware constraint that the addr translation and the data request are integreted into one pipeline ?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hi sir:
I found that the when a load is scheduled because of it having partitial address overlapped with a older store,
it will be resshduled by IEW::rescheduleMemInst --> InstructionQueue::rescheduleMemInst where eventully set the translationCompleted flag and translationStated flag to false
then the load when be issue again will have to do address translation one more time while it has done the translation completely
so why should be this flow? is it be better that just re-do the data request by the physical address having translated before ?
or is this a hardware constraint that the addr translation and the data request are integreted into one pipeline ?
Beta Was this translation helpful? Give feedback.
All reactions