----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2341/#review5279 -----------------------------------------------------------
I think we need better explanation in the opening comment. While I sort of understand what the purpose, I feel lost in the mass of code in cache_impl.hh. src/mem/cache/cache_impl.hh <http://reviews.gem5.org/r/2341/#comment4813> I would not want to call this request WriteInvalidate if it is not writing to the block. Why not just call it invalidate? src/mem/cache/cache_impl.hh <http://reviews.gem5.org/r/2341/#comment4814> I do not like this check. You have marked that WriteInvalidateReq requires a response in packet.cc. And here you are testing that pkt->cmd is not WriteInvalidateReq. - Nilay Vaish On Aug. 15, 2014, 2:24 p.m., Andreas Hansson wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/2341/ > ----------------------------------------------------------- > > (Updated Aug. 15, 2014, 2:24 p.m.) > > > Review request for Default. > > > Repository: gem5 > > > Description > ------- > > Changeset 10308:6bde9d243535 > --------------------------- > mem: write streaming support via WriteInvalidate promotion > > Support full-block writes directly rather than requiring RMW: > * a cache line is allocated in the cache upon receipt of a > WriteInvalidateReq, not the WriteInvalidateResp. > * only top-level caches allocate the line; the others just pass > the request along and invalidate as necessary. > * to close a timing window between the *Req and the *Resp, a new > metadata bit tracks whether another cache has read a copy of > the new line before the writeback to memory. > > > Diffs > ----- > > src/mem/cache/blk.hh 79fde1c67ed8 > src/mem/cache/cache.hh 79fde1c67ed8 > src/mem/cache/cache_impl.hh 79fde1c67ed8 > src/mem/packet.hh 79fde1c67ed8 > src/mem/packet.cc 79fde1c67ed8 > > Diff: http://reviews.gem5.org/r/2341/diff/ > > > Testing > ------- > > > Thanks, > > Andreas Hansson > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
