> On Sept. 23, 2015, 5:12 p.m., Nilay Vaish wrote: > > Is this really snooping? The more I read this code, the more it feels > > that snoop filter is really a directory and that it should be rechristened.
I'd say so :-). The guys at AMD would probably call it a ProbeFilter rather. The SnoopFilter only steers snoops based on its own local crossbar, so there is no SnoopFilter that can tell you what global cache has a block in what state. The latter is what I would consider a directory. Agreed? - Andreas ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/3060/#review7255 ----------------------------------------------------------- On Aug. 21, 2015, 3:49 p.m., Andreas Hansson wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/3060/ > ----------------------------------------------------------- > > (Updated Aug. 21, 2015, 3:49 p.m.) > > > Review request for Default. > > > Repository: gem5 > > > Description > ------- > > Changeset 11063:c64bbbe67e13 > --------------------------- > mem: Add snoop filters to L2 crossbars, and check size > > This patch adds a snoop filter to the L2XBar. For now we refrain from > globally adding a snoop filter to the SystemXBar, since the latter is > also used in systems without caches. In scenarios without caches the > snoop filter will not see any writeback/clean evicts from the CPU > ports, despite the fact that they are snooping. To avoid inadvertent > use of the snoop filter in these cases we leave it out for now. > > A size check is added to the snoop filter, merely to ensure it does > not grow beyond the total capacity of the caches above it. The size > has to be set manually, and a value of 16 MByte is choosen as suitably > high default. > > > Diffs > ----- > > src/mem/XBar.py 842f56345a42 > src/mem/snoop_filter.hh 842f56345a42 > src/mem/snoop_filter.cc 842f56345a42 > > Diff: http://reviews.gem5.org/r/3060/diff/ > > > Testing > ------- > > > Thanks, > > Andreas Hansson > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
