LRU is not at all related to totalrefs. LRU tracks the least recently and least frequently for which u need the total reference. Gem5 correctly models the LRU. It tracks most recently used and moving it to head of the block. For LRU the total refs doen't matter.
One thing that gem5 doesnot as far as replacement is concerend, it's not evicting the invalid lines. In general the policy should try to evict invalid lines if any else the LRU block. On Wed, Jan 4, 2012 at 10:54 PM, Nilay Vaish <[email protected]> wrote: > On Wed, 4 Jan 2012, Mahmood Naderan wrote: > > Why? >>> >> >> Back to my example, I stated: >> >> init >> (b1,0) (b2,0) (b3,0) (b4,0) and totalRefs=0 >> >> b1 is accessed >> (b1,1) (b2,0) (b3,0) (b4,0) >> >> b3 is accessed >> (b3,1) (b1,1) (b2,0) (b4,0) >> >> b1 is accessed >> (b1,2) (b3,1) (b2,0) (b4,0) >> >> b4 is accessed >> (b4,1) (b1,2) (b3,1) (b2,0) >> >> b2 is accessed >> (b2,1) (b4,1) (b1,2) (b3,1) >> >> >> If the above procedure is exactly what is done in gem5, then: >> >> 1) gem5 return totalRefs=1 because >> totalRefs += blk->refCount >> totalRefs = 0 + 1 (from b3) = 1 >> >> 2) based on the description, "total number of references to valid >> blocks", I think is 5 >> >> >> >> > That's because you are looking at the value of totalRefs while the > simulation is running. At the end of the simulation, references for blocks > that are in the cache are added to totalRefs. > > > -- > Nilay > ______________________________**_________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/**mailman/listinfo/gem5-users<http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users> > -- *thanks®ards * *BISWABANDAN*
_______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
