> On Dec. 24, 2015, 3:27 a.m., Steve Reinhardt wrote: > > src/mem/cache/cache.cc, line 1906 > > <http://reviews.gem5.org/r/3255/diff/2/?file=52292#file52292line1906> > > > > could we (at least theoretically) have an uncached write that > > invalidates cached copies as a side effect?
We will eventually need to handle this case. At the moment I'd prefer to know when it happens, hence the panic. For a whole-line write it's "easy", but for partial-line writes we need the CoherentXBar to help out and force writebacks (or do some pretty horrible in-place update). Patches are on the way. - Andreas ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/3255/#review7745 ----------------------------------------------------------- On Dec. 22, 2015, 4:24 p.m., Andreas Hansson wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/3255/ > ----------------------------------------------------------- > > (Updated Dec. 22, 2015, 4:24 p.m.) > > > Review request for Default. > > > Repository: gem5 > > > Description > ------- > > Changeset 11281:84339ca32391 > --------------------------- > mem: Do not alter cache block state on uncacheable snoops > > This patch ensures we do not respond with a Modified (dirty and > writable) line if the request is uncacheable, and that the cache > responding retains the line without modifying the state (even if > responding). > > > Diffs > ----- > > src/mem/cache/cache.cc d9a0136ab8cc > > Diff: http://reviews.gem5.org/r/3255/diff/ > > > Testing > ------- > > > Thanks, > > Andreas Hansson > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
