> On June 18, 2012, 7:23 p.m., Ali Saidi wrote: > > src/mem/cache/tags/cacheset.cc, line 90 > > <http://reviews.gem5.org/r/1275/diff/1/?file=27488#file27488line90> > > > > There must be a way to do this without the loop.
It's just copied from moveToHead(), so I don't think it's fair to make Lena fix it. Also, the LRU list is just stored as an array, so you do have to shift pointers (hence the loop). Both versions could stand to be streamlined a bit though, I agree. - Steve ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/1275/#review2959 ----------------------------------------------------------- On June 18, 2012, 4:17 p.m., Lena Olson wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/1275/ > ----------------------------------------------------------- > > (Updated June 18, 2012, 4:17 p.m.) > > > Review request for Default. > > > Description > ------- > > Changeset 9071:7526dad15b35 > --------------------------- > Cache: Fix the LRU policy for classic memory hierarchy > > The LRU policy always evicted the least recently touched way, even if it > contained valid data and another way was invalid, as can happen if a block has > been invalidated by coherance. This can result in caches never warming up > even > though they are replacing blocks. This modifies the LRU policy to move blocks > to LRU position on invalidation. > > > Diffs > ----- > > src/mem/cache/tags/cacheset.hh fa77985a87c6ff77284ed9bdd4ae304b2e373666 > src/mem/cache/tags/cacheset.cc fa77985a87c6ff77284ed9bdd4ae304b2e373666 > src/mem/cache/tags/lru.cc fa77985a87c6ff77284ed9bdd4ae304b2e373666 > > Diff: http://reviews.gem5.org/r/1275/diff/ > > > Testing > ------- > > > Thanks, > > Lena Olson > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
