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
In LSU we are planing to use shift register to keep age order of instructions. Shift registers are linear. Maybe we should consider using the binary trees to keep elements in order? I am not sure if there will be any benefits of such solution.
The only one which come to my mind is the complexity of the random access in trees we have O(log(depth)) critical path. In shift register we have implementation defined critical path -- we don't know how access to the element in the array will be synthesised by yosys.
The text was updated successfully, but these errors were encountered:
In LSU we are planing to use shift register to keep age order of instructions. Shift registers are linear. Maybe we should consider using the binary trees to keep elements in order? I am not sure if there will be any benefits of such solution.
The only one which come to my mind is the complexity of the random access in trees we have
O(log(depth))
critical path. In shift register we have implementation defined critical path -- we don't know how access to the element in the array will be synthesised by yosys.The text was updated successfully, but these errors were encountered: